Most weight lost on GLP-1 drugs returns within a year after stopping

Stopping popular GLP-1 weight-loss drugs may reverse much of their benefit within a year, according to a large modeling analysis, raising questions about how long patients may need to stay on treatment to sustain meaningful weight loss. 

Study: Trajectory of weight regain after cessation of GLP-1 receptor agonists: a systematic review and nonlinear meta-regression. Image credit: Matt Fowler KC/Shutterstock.com

Glucagon-like peptide-1 receptor agonists (GLP-1RAs) can produce significant weight loss in obese individuals. However, a new study published in the journal eClinical Medicine reports that most people regain about 60 % of the weight lost within a year of stopping treatment.

Growing use of GLP-1 drugs highlights long-term weight challenges

Currently, obesity affects over one billion people worldwide, who have higher odds of adverse outcomes like cardiometabolic disease. Weight reduction by 5 % to 10 % typically improves these odds, though for people with severe obesity (for example, BMI ≥35 kg/m²), 15 % to 20 % weight loss is a more meaningful target.

In practice, this degree of weight loss is difficult to achieve with lifestyle modifications, primarily dietary changes and regular moderate physical activity, though these are considered the cornerstone of weight management.

GLP-1RAs have revolutionized the field of weight loss therapy. They typically induce a 15 % to 20 % weight reduction. Yet their cost and adverse effects cause around half of users to stop them within a year. This is linked to weight rebound.

No large-scale study has yet directly followed weight regain for more than one year after discontinuation. Without such data, treatment guidelines may not be optimal for individual users. For instance, the UK’s National Institute for Health and Care Excellence (NICE) advises that the GLP-1RA semaglutide be used for up to two years for weight loss. This raises the question of whether some patients might require longer treatment to sustain weight loss.

The risk of weight regain needs to be addressed, perhaps through individualized dose-tapering strategies, given the increasing popularity of these agents for weight loss and the high discontinuation rates.

The current study used a systematic review and nonlinear meta-regression modeling approach to extrapolate weight loss trajectories beyond the last-reported follow-up time point after drug cessation. The authors claim that it is “the most comprehensive collection of evidence on post-GLP-1RA weight outcomes to date.”

Weight regain after GLP-1 discontinuation 

The study searched multiple databases for randomized controlled trials (RCTs), non-randomized interventional studies, and observational studies examining weight trajectories after discontinuing GLP-1RAs.

The time points at which weight regain was reported varied significantly across studies. For this reason, the investigators modeled weight rebound on a continuous trajectory using available follow-up data from the included studies.

Earlier studies showed a weight-recovery curve that was almost exponential in shape, with rapid weight regain early on, followed by a shallower slope that reached a new steady state over time. In addition to examining exponential changes in weight after GLP-1RA cessation, the researchers also explored reported changes in glycated hemoglobin (HbA1c) levels and systolic blood pressures.

Model predicts most lost weight returns within one year

The authors retrieved 48 relevant studies involving overweight or obese participants. These included GLP-1RA treatment for durations ranging from 10 to 104 weeks, with post-cessation follow-up periods ranging from 4 weeks to up to 2 years. The agents used included liraglutide, semaglutide, tirzepatide, and other GLP-1RAs.

However, the nonlinear meta-regression that generated the primary weight-regain estimates included six randomized controlled trials involving 3,236 participants.

The findings showed that stopping GLP-1RAs led to a predictable weight rebound with a decelerating trajectory.

The model predicted rapid weight regain immediately after cessation, followed by a slowdown and a gradual approach toward a plateau in the months thereafter. One year after stopping the drug, patients had regained 60 % of the weight they had lost, representing 80 % of the total weight expected to return.

Overall, the model estimated that weight regain would stabilize at about 75 % of the total weight lost during treatment. Patients regained weight by half roughly every 23 weeks, with broadly similar modeled trajectories across different GLP-1RAs. Notably, even after stopping the drug, some weight loss benefit persisted long-term, albeit much weaker than with continued treatment. The authors suggest that sustained weight loss may therefore require ongoing therapy in some patients.

Despite the regained weight, the study suggests that approximately 25 % of the weight loss acutely stays off over time. Assuming an initial reduction of 15 % to 20 % of baseline weight, the model suggests that 4 % to 5 % of the remaining weight is lost in the long term. While superior to placebo, this falls slightly below the commonly cited 5 % threshold considered clinically meaningful for individuals with a BMI below 35 kg/m², and well below the larger reductions often recommended for those with severe obesity.

HbA1c decreased by 0.5–1.5 percentage points during the treatment period, but half of the reduction recovered by 8–12 weeks after cessation. Despite this rise, HbA1c mostly remained below the initial value at one year. Similar trends were observed for systolic blood pressure, although the authors note that these analyses were exploratory and based on limited, inconsistently reported data.

Three studies did not show weight gain after treatment-induced weight loss, although each used different strategies or reported outcomes differently.

One study used a dose-tapering approach, allowing participants to gradually reduce their GLP-1RA dose and, in some cases, stop the drug entirely without an increase in mean weight. However, the report did not clarify how many participants fully discontinued treatment versus those who simply reduced their dosage while continuing therapy.

In another study, participants were placed on a low-carbohydrate diet after stopping the medication, and the researchers reported that they maintained their weight loss.

The third study reported no overall weight regain but did not provide individual weight outcomes, limiting interpretation of the results.

Separately, another study suggested that longer durations of GLP-1RA treatment may help sustain weight loss even after discontinuation, although this hypothesis remains to be validated.

The studies mostly had a moderate risk of bias, limiting the validity of the conclusions. Moreover, the data used for this model did not extend beyond 52 weeks, necessitating extrapolation. Future research should use longer follow-up periods to validate the model estimates.

The researchers included only studies that reported a mean weight loss of 3 kg with GLP-1RA treatment, to ensure that significant post-cessation weight regain could be captured. This excludes a subgroup with lesser degrees of weight loss, but which is still clinically relevant.

Residual confounding is likely due to other medications and weight-loss strategies used alongside these drugs. These could not be adjusted for due to inadequate reporting in most studies.

Long-term treatment strategies may be needed to sustain weight loss

These considerations underscore the importance of a flexible, individualized approach to weight management when using GLP-1RAs.

These agents may need to be used for longer durations or discontinued using tailored strategies to maintain effective weight loss. This is especially important as newer and more potent drugs are on the horizon for obesity management.

Download your PDF copy by clicking here.