Stopping semaglutide or tirzepatide: weight regain and tapering

The original withdrawal studies set the expectation. In the STEP 1 extension, people who stopped semaglutide 2.4 mg after 68 weeks regained about two thirds of their lost weight within a year, and the cardiometabolic improvements reversed with it. In SURMOUNT-4, people switched from tirzepatide to placebo after 36 weeks regained about 14 percent of body weight over the following year while those who stayed on lost a further 5 percent. Both trials stopped the drug in one step, with no taper. They describe a fast curve, steepest in the first few months, and the 2026 studies add a wider view across drugs and across people who were not in a trial.

The Oxford review, published in January 2026, pooled 37 studies of stopping obesity medication. Across all drugs, weight came back at about 0.4 kg a month on average. After semaglutide or tirzepatide the rate was about 0.8 kg a month, roughly double, which fits the fact that those drugs produced the most loss and so had the most to give back. Projecting the curve forward put the average return to baseline at roughly 1.5 years.

Two details are easy to miss. Regain was faster right after stopping and slowed over time, so the monthly average understates the first quarter and overstates the second year. And the projection to baseline is a projection: most included studies did not follow people for 18 months, and the few that did show the curve flattening.

In March 2026, Cleveland Clinic published a real world analysis of nearly 8,000 patients who had stopped a GLP-1 drug. Average regain was lower than the trial and review figures. The reason is in the design. A real world cohort includes people who stopped for a few months and restarted, people who switched to another drug, and people who stopped at a low dose after modest loss, none of whom appear in a trial where the drug is withheld. Average regain across that mix is smaller because much of the mix went back on treatment.

The two findings are not in conflict. The review measures what happens if you stop and stay stopped. The Cleveland Clinic data measure what people actually do, which is often to stop and then not stay stopped.

The cohort figure that circulates most is from a 24 month follow up that counted full regain, meaning a return to or above the starting weight. It reported that in 23 percent of semaglutide users and 21 percent of tirzepatide users at two years. Read one way, a fifth of people lost everything they had gained. Read the other way, roughly four in five were still below their starting weight two years after stopping.

The drug leaves. Semaglutide has a half-life of about 7 days and is largely cleared about 5 to 7 weeks after the last dose; tirzepatide's half-life is about 5 days and it is gone in about 4 weeks. Appetite suppression fades over those weeks, and the slowed gastric emptying that kept meals feeling large fades with it. People describe food noise returning in the second to fourth week.

The body it leaves behind is also different from the body that started. Weight loss by any method lowers leptin, raises ghrelin and reduces resting energy expenditure, and those shifts persist for at least a year in the classic diet studies. The drug was holding appetite below where a lighter body would normally set it. Remove it and the set point pulls back, in a smaller body burning fewer calories than at the start. More hunger and less expenditure is why regain is the norm, and the same regain follows diets and every other intervention short of surgery.

No controlled trial has compared tapering a GLP-1 drug against stopping it in one step. The withdrawal trials stopped abruptly, and their regain curves are the only randomized data. The compound pages describe stopping as a gradual taper, and that reflects clinical practice and the pharmacology rather than a trial result. A slower reduction gives appetite time to return gradually and the person time to adjust intake.

The counter argument is that a taper only delays the same endpoint. A dose of zero is a dose of zero whether reached in one week or three months, and if appetite is set by the drug level, the regain simply starts later. Which view is right is unknown. Trials testing step down schedules and reduced maintenance doses are ongoing, including Lilly's SURMOUNT-MAINTENANCE study, with results expected late 2026. Until one reports, tapering is a reasonable preference with no evidence that it changes the outcome.

What a taper does do is reveal the dose at which appetite control is lost. Someone who steps down from 2.4 mg to 1 mg of semaglutide and holds weight for three months has learned something a cold stop would not have shown.

The alternative to stopping is not stopping. The labels for both drugs describe chronic use, and the maintenance doses studied were the full ones: semaglutide 2.4 mg, tirzepatide 5, 10 or 15 mg. Many people hold at a lower step that controls appetite, which is within the label for both drugs since increases are only as tolerated. That is maintenance on a labeled dose, and the semaglutide dosage chart and tirzepatide dosage chart show where each dose sits on the ladder.

Below the labeled steps is the territory covered in GLP-1 microdosing: holding at 0.25 or 0.5 mg of semaglutide or at 2.5 mg of tirzepatide after reaching a goal, or going lower still. The idea is that a smaller signal might hold weight that a full dose was needed to lose. No trial supports it yet, and since zero dose removes the effect, a reduced dose should be expected to produce a reduced effect. SURMOUNT-MAINTENANCE will test the hypothesis properly, for tirzepatide. A taper and a maintenance microdose can look identical from the outside; the difference is intent and what happens if weight starts moving.

The strategies with evidence are the ones that work after any weight loss. Protein in the range of 1.2 to 1.6 grams per kilogram a day, which preserves lean mass and is the most satiating macronutrient per calorie. Resistance training two or three times a week, which protects the muscle that sets resting expenditure. Regular self weighing, which catches a 2 kg drift before it becomes 10. These are the counters covered in muscle loss on GLP-1 drugs, and they matter more once the drug is no longer doing the appetite work.

The things people do without evidence are worth naming too. Switching to a grey market compound such as retatrutide at a low dose is a change of drug, not a maintenance strategy, and carries every purity and supervision problem the buying guide describes. Adding growth hormone secretagogues to preserve muscle has no human trial behind it. Restarting if weight returns is what the Cleveland Clinic data suggest most people do, and it means a fresh titration with fresh side effects.

The American Diabetes Association's 2026 Standards of Care recommend continuing obesity medication through weight maintenance rather than stopping once a goal weight is reached. The reasoning is the data above: the drugs treat a chronic condition, the benefit depends on continued use, and stopping reverses it, including the blood pressure, lipid and glucose gains the STEP 1 extension saw drift back with the weight.

For someone deciding whether to stop, the honest framing is a trade: the cost of continuing, in money, side effects and long term uncertainty, against a likely regain of most of the weight over a year or two. Cost is covered in what peptide therapy costs. A clinician who knows the person's history is the right place to work through that trade.