Eating Habits and Their Impact on T2D Treatments

Studies have shown that emotional eating and other eating habits and behaviors could reduce weight loss when taking GLP-1 receptor agonists.

Emotional eating is eating to respond to negative emotions while also ignoring internal physiological hunger signals. Emotional eating scores have been linked to obesity, less weight loss after weight loss programs and weight regain after an intervention. Emotional eating is also associated with altered brain signaling in reward processing areas. Furthermore, obesity and type 2 diabetes mellitus are linked to changes in reward processing areas.

Obesity and type 2 diabetes mellitus treatment option GLP-1 receptor agonists (GLP-1RAs) are effective due to suppressing appetite signaling in the brain. Past studies have found that emotional eating is associated with less sensitivity to GLP-1 receptor activation and the effects of appetite signaling in the brain after short-term treatment. However, the following eating styles have yet to be assessed: restraint eating (cognitive control of food intake to control body weight) and external eating (overeating due to external food-related cues, such as sight and smell). In this study, researchers investigated if people with type 2 diabetes with various eating styles are less sensitive to GLP-1RA ‘s after long-term treatment.

This open-label randomized crossover intervention study was a secondary analysis of the LIBRA study. The LIBRA study aimed to assess the effect of liraglutide compared to insulin glargine on brain activation in response to food cues using functional magnetic resonance imaging. This secondary analysis created two treatment periods for 12 weeks and a washout period of 12 weeks in between. In one treatment period, patients received liraglutide at escalating doses each week until reaching a final dose of 1.8 mg daily. Patients received insulin glargine at 10 IU daily in the other treatment period. The dose was increased based on fasting self–monitored blood glucose levels. Three testing visits were done for each treatment period at baseline, ten days, and 12 weeks. Patients had functional magnetic resonance imaging scans done at fasting and postprandial conditions. Visual food cues were additionally presented to the patients. At the end of the visit, patients were presented with a food buffet where their energy intake was monitored. The Dutch Eating Behavior Questionnaire (DEBQ) was used to determine baseline emotional, external, and restraint eating scores. The longitudinal differences between treatments were analyzed using a generalized estimating equation approach. Pearson regression coefficients were used to make associations between eating behaviors and differences in brain activation. Results were significant at P <0.05. The patient population consisted of Caucasian men and postmenopausal women aged 40 – 65 with type 2 diabetes. Patients had an HbA1c between 6.0% and 8.5%, a body mass index >26 kg/m², and were treated with metformin with or without sulfonylurea derivatives. Exclusion criteria included patients with a history of using any central acting agent or glucocorticoids.

In the fasting state, baseline emotional eating scores were negatively associated with liraglutide–induced reductions in the left caudate (p = 0.055) in response to high–calorie versus non–food pictures after 12 weeks of treatment. Similar results were found in the postprandial condition in the left caudate (p = 0.055). However, baseline emotional eating scores were not associated with brain responses after receiving chocolate milk versus tasteless solution after 12 weeks of treatment.

Baseline external eating scores had a negative correlation with liraglutide – induced reductions in the fasting condition, in comparison to insulin glargine in response to food versus non – food pictures in the right insula (r = – 0.63, p = 0.047), after 10 day treatment. Baseline eating scores had a negative correlation to liraglutide-induced reductions in the postprandial condition, in the right insula (r = – 0.77, p = 0.003) in response to food versus non-food pictures, and in the bilateral amygdala (right, r = – 0.–0.54, p = 0.043; left, r = – 0.68 p = 0.005), and right insula (r =  0.67, p = 0.026) after 10 day treatment. There were no associations between external eating scores and brain responses after 12 weeks of treatment.

Baseline restraint eating scores had a positive association with liraglutide-induced increases in comparison to insulin glargine in the right putamen (p = 0.084) and right insula (p = 0.060) after receiving chocolate milk versus tasteless solution, after 10 day treatment. Baseline restraint eating scores had a positive association with liraglutide-induced reductions in the fasting condition, in the right and left insula (p = 0.049, r = 0.61; p = 0.001, r = 0.79, respectively) in response to food versus non-food pictures, and in the left insula (p = 0.027, r = 0.64), and right caudate nucleus (p = 0.043, r = 0.60) in response to high-calorie versus non-food pictures, after 12 week treatment.

This study found that at baseline, higher emotional eating scores were correlated to less pronounced GLP – 1RA induced changes in the brain’s response to receiving food and food pictures amongst individuals with type 2 diabetes after ten days of treatment. This was also seen to a lesser extent after 12 weeks of treatment. Additionally, baseline external eating scores were associated with less GLP-1 RA-induced changes in brain response after ten treatment days. However, higher baseline restraint eating scores were correlated with GLP-1RA induced changes in brain responses after 12-week treatment. This data suggests that elements of emotional eating make individuals less sensitive to GLP 1-RA treatment.

Furthermore, external eating may also decrease, whereas restrained eating may increase the patient’s sensitivity to the effects of GLP-1 RA’s. These results can explain why GLP-1RA’s are not associated with considerable weight loss in all patients. Therefore, future studies should be done in more extensive and more diverse patient populations for more extended periods to continue examining the impact of eating behavior on the effects of GLP-1 RA’s.

Practice Pearls:

• Emotional eating habits led to a decrease in the effects of GLP-1 RA’s in patients with type 2 diabetes.
• Eating behaviors / habits can play a role in weight loss when a patient takes GLP-1 RA’s.
• Eating behaviors should be addressed to optimize treatment strategies for patients with overweight.

van Ruiten, Charlotte C et al. “Eating behavior modulates the sensitivity to the central effects of GLP-1 receptor agonist treatment: a secondary analysis of a randomized trial.” Psychoneuroendocrinology vol. 137 (2022): 105667. doi:10.1016/j.psyneuen.2022.105667

Amanda Roberts, PharmD Candidate, Florida A&M University, College of Pharmacy and Pharmaceutical Sciences Institute of Public Health