Obesity: What happens after stopping incretin mimetics?

In 2022, 38% of the world’s population (2.6 billion people) were either overweight or obese, conditions that increase the risk of morbidity and premature mortality. Incretin mimetics, weight management drugs that include GLP-1 agonists, GIP agonists, and combined GLP-1/GIP agonists, now play a central role in the management of obesity. They have profoundly changed the therapeutic landscape, with body weight losses of around 15 to 20% of initial weight. These results have generated legitimate enthusiasm among both healthcare professionals and patients, offering the prospect of an effective pharmacological alternative where behavioral approaches (diet and physical activity) alone often struggled to produce lasting effects.

But obesity is a chronic disease, marked by powerful biological mechanisms that defend body weight. The key question is therefore not only whether these treatments cause weight loss, but what happens once treatment is stopped. In practice, however, discontinuation of medication is common, whether for reasons of tolerance, cost, availability, or long-term adherence. It is estimated that one in two people stop this treatment within a year. Understanding weight dynamics after stopping these treatments, and how they compare to those observed after stopping conventional behavioral programs, is therefore essential to assessing their true long-term impact on health.

The study

To this end, researchers at the University of Oxford conducted a systematic review and meta-analysis of all available studies that evaluated changes in body weight after discontinuing drug treatment in overweight or obese adults. Major biomedical databases were searched from their inception to early 2025, with no date or language restrictions, to identify relevant randomized trials, non-randomized trials, and observational studies.

To be included, studies had to include a drug treatment phase of at least eight weeks, followed by a drug-free follow-up period of at least four weeks. The molecules studied covered a broad spectrum, ranging from older treatments such as orlistat and sibutramine to newer incretin mimetics such as semaglutide and tirzepatide. A total of 37 studies, representing 63 intervention arms and 9,341 participants, were included in the analysis. Of these studies, 35 were randomized controlled trials and 28 studies had a control group during both the intervention and follow-up periods.

The authors used several complementary statistical models to estimate the trajectory of weight regain over time, expressing the results as average monthly rates. They also analyzed changes in cardiometabolic markers after discontinuation of treatment and compared these dynamics to those observed after discontinuation of behavioral weight management programs, based on previous work conducted by the same team.

Results & Analysis

The main results show that discontinuation of incretin-mimetic drugs is followed, on average, by rapid and progressive weight regain, with an average rate of 0.4 kg per month. This regain occurs regardless of the type of drug used, including the most recent and effective molecules during the treatment phase. The observed trajectory suggests that body weight tends to return to its initial value within 1.7 years after stopping treatment.

But this dynamic does not only apply to weight. The cardiometabolic improvements achieved during the weight loss phase follow a parallel trajectory. Thus, after stopping treatment, glycated hemoglobin (HbA_1C), fasting blood glucose, cholesterol, triglycerides, and blood pressure (diastolic and systolic) gradually return to their baseline values within 1.4 years, suggesting that cardiometabolic benefits are closely dependent on maintaining the weight loss induced by the drug.

The study authors also point out that there was insufficient data to analyze changes in cardiometabolic markers after treatment with newer and more effective incretin mimetics (only one study with a 2-year follow-up after stopping treatment with semaglutide, for example), but they still show that weight regain is faster (0.8 kg/month) and that a return to initial weight is expected 1.5 years after stopping treatment, implying that cardiovascular health benefits will also likely diminish more quickly.

One particularly interesting element is the comparison with behavioral programs. Although drugs lead to greater weight loss on average, more quickly, weight regain after stopping treatment is faster than that observed after the end of programs based on diet and physical activity. In addition, cardiometabolic benefits last at least 5 years after stopping the program. This difference persists even after adjusting for the extent of initial weight loss, suggesting that the nature of the intervention, and not just the weight lost, influences the ability to maintain benefits over time.

These results are consistent with physiological logic. Weight-loss drugs work primarily by altering hunger and satiety signals, making eating behavior easier to control as long as the treatment is present. Once this pharmacological support is withdrawn, biological weight defense mechanisms take over in the absence of sustainably integrated behavioral strategies.

Practical applications

These data invite broader reflection on the role of drugs in the management of obesity. They do not call into question their short-term effectiveness or their value in certain clinical situations. However, they do highlight the limitations of occasional or short-term use without an overall strategy aimed at maintaining long-term adaptations. As obesity is a chronic and recurrent condition, prolonged treatment with incretin mimetics may be necessary to maintain health benefits. One trial showed that continuous treatment with semaglutide maintained weight loss for four years. However, in the United States and Denmark, early data show that dropout rates outside of clinical trials are approximately 50% at one year. These data suggest that despite their success in initial weight loss, these drugs alone may not be sufficient for long-term weight control.

Furthermore, these results serve as a reminder that drug-induced weight loss does not constitute a lasting “reset” of the weight regulation system. Without prolonged support, consolidated behavioral changes, and consideration of the actual duration of treatment, the likelihood of a rebound effect is high. This does not mean that these treatments are useless, but that they should be considered as one tool among others, not as a stand-alone solution.

For healthcare professionals, the findings of this study reinforce the importance of discussing what will happen after treatment begins. For patients, they help set realistic expectations: the drug facilitates weight loss, but it does not replace the fundamental work of changing lifestyle habits or the need for long-term follow-up. The study authors also explain that a survey of American adults revealed that 45% of them were interested in using GLP-1 receptor agonists for weight management, but this proportion drops to 14% when people are informed about weight regain after stopping treatment.

Référence

West S, Scragg J, Aveyard P, Oke JL, Willis L, Haffner SJP, Knight H, Wang D, Morrow S, Heath L, Jebb SA & Koutoukidis DA. Weight regain after cessation of medication for weight management: systematic review and meta-analysis. BMJ 2026, 392 : e085304.