What the First Evidence Actually Establishes
The post-cessation GLP-1RA trajectory now has two useful lenses: a one-year semaglutide withdrawal case and a broader review that models how regain changes over time. Connecting them supports a practical judgment about the evidence: substantial regain after treatment stops is a recurring population pattern, but its final level, causes, and best response remain less certain. That distinction is more useful than either a reassuring claim that the loss persists or a fatalistic claim that everyone returns to baseline.
The central finding is not a single percentage. It is a relationship between a treatment state and a withdrawal state. During treatment, participants experienced a large reduction in weight. After treatment ended, many moved upward again, with the available pooled data suggesting more rapid change early and slower change later. The amount of movement must still be interpreted against the amount lost, the length of observation, and the conditions under which withdrawal occurred.
The common pattern is substantial, decelerating rebound
The broad review collected 48 studies of adults with overweight or obesity who stopped a GLP-1RA. Almost all reported some degree of weight regain. In the subset with more than 3 kilograms of initial treatment loss, regain at or around 12 weeks ranged from approximately 10% to 80% of the treatment-induced loss. The range is wide, but the direction is not evenly divided between regain and no regain. The repeated observation is that weight tends to move upward after cessation, while the amount and timing differ.
The fitted trajectory adds shape to that observation. Across six better-documented randomized trials involving 3,236 participants, the model estimated about 60% of the original treatment loss regained by 52 weeks. It projected a maximum of about 75.3% of the loss, with a confidence interval of 68.9% to 81.6%. The curve rises more quickly at first and then slows. This does not mean that every participant followed a smooth exponential path. It means that a changing-rate curve summarized the available population observations better than a single fixed regain value.
The STEP 1 extension makes the same relationship concrete. After 68 weeks of semaglutide and lifestyle counseling, selected treatment completers were observed for another 52 weeks without active study treatment or structured lifestyle support. The semaglutide group had lost 17.3% of body weight during the treatment phase and regained 11.6 percentage points by week 120, leaving a net mean loss of 5.6% from baseline. The study described that regain as roughly two-thirds of the prior loss. The case and the pooled model are not identical estimates, but they point in the same direction: a large part of the treatment response can be reversed within a year of withdrawal.
That conclusion does not imply a universal return to the starting weight. In STEP 1, 48.2% of participants in the semaglutide arm still had at least 5% weight loss from baseline at week 120, although this was down from 86.4% at the end of active treatment. The model also projected a long-term level below baseline.