Early metabolomic, lipid and lipoprotein changes in response to medical and surgical therapeutic approaches to obesity

BACKGROUND: Glucagon-like peptide-1 receptor agonists (GLP-1RA) and bariatric surgery have proven to be effective treatments for obesity and cardiometabolic conditions. We aimed to explore the early metabolomic changes in response to GLP-1RA (liraglutide) therapy vs. placebo and in comparison to bariatric surgery. METHODS: Three clinical studies were conducted: a bariatric surgery cohort study of participants with morbid obesity who underwent either Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG) studied over four and twelve weeks, and two randomized placebo-controlled, crossover double blind studies of liraglutide vs. placebo administration in participants with type 2 diabetes (T2D) and participants with obesity studied for three and five weeks, respectively. Nuclear magnetic resonance spectroscopy-derived metabolomic data were assessed in all eligible participants who completed all the scheduled in-clinic visits. The primary outcome of the study was to explore the changes of the metabolome among participants with obesity with and without T2D receiving the GLP-1RA liraglutide vs. placebo and participants with obesity undergoing bariatric surgery during the three to five-week study period. In addition, we assessed the bariatric surgery effects longitudinally over the twelve weeks of the study and the differences between the bariatric surgery subgroups on the metabolome. The trials are registered with ClinicalTrials.gov, numbers NCT03851874, NCT01562678 and NCT02944500. RESULTS: Bariatric surgery had a more pronounced effect on weight and body mass index reduction (-14.19 ± 5.27 kg and - 5.19 ± 5.27, respectively, p < 0.001 for both) and resulted in more pronounced metabolomic and lipidomic changes compared to liraglutide therapy at four weeks postoperatively. Significant changes were observed in lipoprotein parameters, inflammatory markers, ketone bodies, citrate, and branched-chain amino acids after the first three to five weeks of intervention. After adjusting for the amount of weight loss, a significant difference among the study groups remained only for acetoacetate, β-hydroxybutyrate, and citrate (p < 0.05 after FDR correction). Glucose levels were significantly reduced in all intervention groups but mainly in the T2D group receiving GLP-1RA treatment. After adjusting for weight loss, only glucose levels remained significant (p = 0.001 after FDR correction), mainly due to the glucose change in the T2D group receiving GLP-1RA. Similar results with those observed at four weeks were observed in the surgical group when delta changes at twelve weeks were assessed. Comparing the two types of bariatric surgery, an intervention effect was more pronounced in the RYGB subgroup regarding total triglycerides, triglyceride-rich lipoprotein size, and trimethylamine-N-oxide (p for intervention: 0.031, 0.028, 0.036, respectively). However, after applying FDR correction, these changes deemed to be only suggestive; only time effects remained significant with no significant changes persisting in relation to the types of bariatric surgery. CONCLUSIONS: The results of this study suggest that the early metabolomic, lipid and lipoprotein changes observed between liraglutide treatment and bariatric surgery are similar and result largely from the changes in patients' body weight. Specific changes observed in the short-term post-surgical period between bariatric vs. nonsurgical treated participants, i.e., acetoacetate, β-hydroxybutyrate, and citrate changes, may reflect changes in patient diets and calorie intake indicating potential calorie and diet-driven metabolomics/lipidomic effects in the short-term postoperatively. Significant differences observed between SG and RYGB need to be confirmed and extended by future studies.