SURMOUNT-MAINTAIN results published: staying on the full dose held all the weight loss

Eli Lilly's SURMOUNT-MAINTAIN trial, published in The Lancet on May 12, 2026, is the first randomized test of lowering the tirzepatide dose instead of stopping it. After 60 weeks of open-label treatment at the maximum tolerated dose, participants who stayed on that dose held onto essentially all of their weight loss for another year, those who dropped to 5 mg gave some back, and those switched to placebo regained the most ^[1][3].

What the trial did

The phase 3b trial ran across 20 US sites and enrolled 441 adults with a BMI of 30 or higher, or 27 or higher with a weight-related condition, who had at least one unsuccessful dieting attempt ^[1]. Everyone first took once-weekly tirzepatide (sold as Zepbound for weight management) for 60 weeks, escalating to a maximum tolerated dose of 10 mg or 15 mg ^[1].

At week 60, the 378 participants who had lost at least 5% of their body weight and tolerated the drug were randomly assigned 3:3:2 to continue at their maximum tolerated dose (140 people), drop to 5 mg (144), or switch to placebo (94) for 52 more weeks ^[1][3]. Baseline average body weight was 113.8 kg, average BMI was 40.1, mean age was 46.6 years, 65% were women and 67% were White ^[1]. Median duration of obesity was 13 years ^[3]. Of the 378, 345 (91%) finished the study ^[1].

The numbers

At week 112, total weight change from the original starting point was −21.9% (95% CI −23.5 to −20.3) for the maintained dose, −16.6% (95% CI −18.0 to −15.1) for 5 mg, and −9.9% (95% CI −11.1 to −8.8) for placebo ^[1]. Both tirzepatide groups beat placebo, with treatment differences of −12.0% and −6.6% (p<0.0001 for all comparisons) ^[1][3].

In Lilly's terms, people continuing the maximum tolerated dose preserved all of their prior weight loss on average, while the 5 mg group regained an average of 5.6 kg (12.3 lb) of what they had lost ^[2]. Lilly also reported figures from a second statistical approach called the efficacy estimand, where the maintained-dose group was down 25.2 kg (22.4%) and the 5 mg group down 19.2 kg (17.0%) from baseline ^[2]. The slightly different percentages reflect different analysis methods, not different people ^[1][2].

Starting at week 84, participants who regained more than half their lost weight could get rescue tirzepatide. That happened for 11 of 138 (8%) in the maintained-dose group, 35 of 142 (25%) at 5 mg, and 60 of 90 (67%) on placebo ^[1].

During the maintenance year, the most common side effects with the maintained dose and 5 mg versus placebo were diarrhea (7.2% and 4.9% vs 1.1%), vomiting (6.5% and 0.7% vs 0%) and nausea (5.8% and 4.2% vs 2.2%) ^[2]. Dropouts due to side effects in that period were 0%, 0.7% and 0% ^[2]. Gastrointestinal events were mostly mild to moderate and mostly happened during dose escalation ^[1].

Why it matters for patients

This trial addresses a question many people on these drugs raise with their clinicians: whether a smaller dose can hold results once the weight comes off. The authors concluded that reducing to 5 mg "might provide a valuable alternative to discontinuation, although individuals' treatment response might vary" ^[1]. It is not equivalent to staying at the top dose — the 5 mg group ended about 5 percentage points higher in body weight and triple the share needed rescue therapy ^[1].

One limit matters for the online conversation about "microdosing." The lowest dose tested was 5 mg, a standard tirzepatide dose, not a fraction of one ^[1]. The trial says nothing about smaller, unapproved amounts. It also excluded people with type 2 diabetes and ran at only 20 US sites ^[1][2].

An accompanying Lancet editorial by André J. Scheen noted the rationale for maintenance at the lowest effective dose had been acknowledged but lacked trial evidence until now ^[3].

What happens next

The trial is complete, having enrolled from Sept 20, 2023 to Jan 20, 2026 ^[1]. A correction notice appeared in The Lancet on June 6, 2026 ^[1]. Lilly says tirzepatide studies in chronic kidney disease and in obesity-related illness and death are still running ^[2]. Whether regulators or guideline writers will formally address dose reduction for maintenance is not yet known.