In addition to the expected achieving good glycemic control, this study demonstrated that treatment with liraglutide over a 12-week period led to a mean weight loss of 5.03 kg in Chinese obese T2D, only one subject’s weight did not change, and 61.3% of the subjects lost more than 5% of body weight from baseline. We also observed the changes of liraglutide-induced body composition assessed by DXA and CT. It was shown that reductions in body weight with liraglutide primarily came from reductions in fat mass rather than lean mass, and decreases in VAT area seemed greater than in SAT. The most interesting findings reported here were the significant correlations between weight loss and increases in both plasma ANP and BNP levels following 12-week liraglutide treatment. We also found that NPs increased particularly more in patients who lost weight more than 5% compared with patients who lost weight less than 5%, suggesting the possible role of increases in circulation NPs levels on the liraglutide-induced weight loss. Given the “rediscovery” of brown adipocytes in adult humans[24, 25] and the potential role of ANP and BNP to stimulate the appearance of brown-like adipocytes associated with increased expenditure, our results imply that increases in plasma NPs may add another dimension to explain how liraglutide induces sustained weight loss.
Liraglutide-induced weight loss might be associated with increases in plasma NPs concentrations
The LEAD-2 and LEAD-3 studies have shown that liraglutide (as monotherapy or in combination with metformin) resulted in greater reductions in fat mass (1.0–2.4 kg) than in lean mass (1.5 kg), confirming our results in which fat mass decreased 3.8 kg and lean mass decreased 1.5 kg after 12-week liraglutide treatment. In addition, the results from CT assessment were also in line with the results from LEAD-2 shown reductions in VAT area were larger than SAT. These are all desirable traits because increased fat mass are associated with decreased insulin sensitivity and increased morbidity and mortality, furthermore, visceral fat is causally associated with insulin resistance. In this study, we found that there was a significant correlation between change in total body fat and change in HOMA-IR. Noticeable, the reductions in fat mass in our study were greater than those of previous studies in terms of the changes in liraglutide-induced body composition, including the LEAD-2, LEAD-3 studies, as well as a shorter study in which fat mass decreased 1.0 kg after 8-week treatment with liraglutide 0.6 mg once-daily. What are the reasons for the excellent results reported in our study? It might be because body fat distribution differs across ethnic background, Chinese and South Asian cohorts had relatively greater amount of abdominal adipose tissue, and this difference was more pronounced in VAT. It is possible that liraglutide exhibited its utmost effectiveness on weight reduction merely due to genetic differences, confirming the hypothesis of one recent study from Japanese in which liraglutide administrated 0.9 mg once-daily achieved about 10% of body weight reduction after 6 months, the result was also much greater than those mainly done in Caucasians[13, 29], however the Japanese study failed to observe the reductions in body fat tissue, and the present study using DXA and CT to assess the changes in liraglutide-induced body compositions only performed in Chinese obese T2D. Therefore, the excellent results of liraglutide-induced reductions in fat tissue should be supported in the future in a double-blind placebo-controlled clinical trial across ethnic groups.
Lowering of energy intake has been shown with liraglutide as well as native GLP-1 both in animals[30, 31] and human subjects. In rats, intracerebroventricular administration of GLP-1 has been shown to induce a marked reduction in food intake, presumably by interacting with GLP-1 receptors, which are present in several areas in the central nervous system[32, 33]. In pharmacological doses, exogenous administration of GLP-1 has been shown to slow gastric emptying substantially. In clinical studies, it was demonstrated that appetite, food preference and eating behavior were significantly changed by liraglutide treatment[34–36]. Therefore, the underlying mechanisms of liraglutide inducing weight loss are partly related to the combined effects of GLP-1 on the gastrointestinal tract and the brain, leading to delaying gastric emptying, inducing satiety and changing eating behavior to reduce the energy intake.
The effect of liraglutide on energy expenditure is a novel observation that requires further investigation[27, 34]. In patients with T2D or not with liraglutide treatment, a weight loss of 2.4 kg ~7.8 kg could be maintained over 52 weeks to 2 years. Since gastrointestinal effects such as vomiting and nauseas were mostly transient, it was apparent that induction of nausea and less energy intake cannot fully explain the liraglutide-induced weight loss in such long term. Recently, it was reported that GLP-1 could increase adipocytes lipolysis in a dose-dependent manner. Moreover, Minsuk Kim et al. have demonstrated that cardiac GLP-1R activation promoted the secretion of atrial natriuretic peptide (ANP). NPs were also shown to increase lipid oxidation and adipocyte thermogenesis, which was subsequently found to promote “browning” of white adipocytes in mouse and human adipocytes through p38 MAPK pathway. In addition, increases in plasma NPs levels were proved to be associated with weight loss induced by gastric bypass surgery and lifestyle changes in human, suggesting that increased NPs play an important role in the weight loss. An attractive hypothesis was raised whether the liraglutide-induced weight loss is associated with the increased circulation NPs levels. In our study, we found that increases in both ANP and BNP levels with liraglutide treatment were associated with reductions in body weight. Moreover, the patients who lost weight more than 5% had approximate 2.5-fold and 3.9-fold intra-individual increases in plasma ANP and BNP levels respectively, compared with the patients who lost weight less than 5% (Figure 2C and Figure 2D). Taken together, our data suggested that liraglutide-induced weight loss primarily from reductions in fat mass might be associated with changes in the NP system.
Possible mechanisms of liraglutide increasing circulating NPs levels
The mechanisms underlying the relative NPs deficiency in obesity are not clear. NP clearance receptors (NPRC) which was found on adipocytes, referred to as the clearance receptor binds ANP and BNP to remove them from circulation, and elevated NPRC have been demonstrated in patients with obesity, suggesting that a putative role for adipose tissue in the clearance of NPs from the circulation[40–42]. This is supported by an animal model of NPRC-/- null mice in which reduced adipose tissue associated with increased NPs has been observed, and the typical brown adipocyte markers were elevated in both BAT and WAT. It is well known that NPs are synthesized and released from the ventricles in response to increased cardiac wall. In this study, we excluded the patients with heart failure at baseline that would account for the observed increase in NPs levels. Moreover, the protective role of GLP-1 in cardiovascular disease has been suggested in both animal experiments and clinical studies. Bao W et al. revealed that the long acting GLP-1 receptor agonist could provid more sustained cadioprotective effect in the setting of acute myocardial I/R injury than the short-acting exendin-4. Clinically, beneficial effects of GLP-1 have also been demonstrated in patients with heart failure. Endogenous GLP-1 level was found to be increased in patients with high cardiovascular risk, suggesting it represent a contra-regulatory response in states of metabolic disorder, hepertriglyceridemia and insulin resistance. In accordance with those previous studies, we observed that treatment with liraglutide could significantly improve cardiovascular risk factors including triglycride, total cholesterol, HDL cholesterol and blood pressure, but not the LDL cholesterol. A study among the Japanese obese patients with T2D also found a significantly reduction in LDL-C with liraglutide at 6 months after discharge. In addition, a recent study demonstrated that liraglutide led to a reduction of blood pressure through activating the secretion of ANP, but this effect was eliminated in the GLP-1R deficient mouse model, suggesting that liraglutide regulate blood pressure via GLP-1R and ANP axis dependent way. We therefore hypothesized that the increases in the NPs levels were not secondary to underlying cardiac changes, but attributed to the liraglutide treatment. With a deeper molecular and clinical understanding of how liraglutide increase NPs levels to regulate fat metabolism and cardiovascular function in the future, the researchers may find new ways to manage obese T2D.
Limitations and conclusion
Several limitations of the present study deserve mention. The study involves a relative small sample size and an observational design without use of control subjects. The future study need to confirm our results, which should be to power a larger, more definitive study incorporating appropriate controls. This might include a patient group treated with an agent that improves glycemic control but not induce weight loss and a second group in whom weight loss alone is achieved (e.g. orlistat) to determine the independent effect on increases in NPs levels following liraglutide treatment. Additionally, the future study should be designed to observe the longitudinal changes of NPs following liraglutide-induced weight loss.
In summary, we found there was a significant correlation between the increases in NPs levels and the reductions in weight loss following the 12-week liraglutide treatment. We also showed that NPs increased particularly more so in patients who lost weight more than 5% compared with patients who lost weight less than 5%. Our data imply that increases in plasma NPs may add another novel dimension to explain how liraglutide induces sustained weight loss. It would seem that a much better understanding of the molecular basis for changes in NPs following the weight loss induced by liraglutide is needed in the future.