Tirzepatide Weight Regain at 3 Years

A new post hoc analysis of the three-year SURMOUNT-1 trial found that 70% of the selected tirzepatide participants regained less than 5% of their starting body weight after reaching their lowest weight. Another 22% regained 5% to less than 10%, while 8% regained at least 10%.^[1] The analysis does not show that regain is harmless. It suggests that a small rise from the lowest recorded weight did not erase most measured metabolic improvements while treatment continued.

This was not a new randomized comparison. Researchers regrouped existing trial participants according to weight regained between their individual nadir and week 176. That makes the paper useful for describing patterns among selected responders, but weaker for proving what caused those patterns or defining a universal threshold for clinically important regain.^[1]

Who was included in the analysis

SURMOUNT-1 was a phase 3, double-blind trial that randomized adults to weekly tirzepatide or placebo. The three-year extension followed 1,032 participants who had obesity or overweight plus prediabetes at baseline. The main trial report found mean weight changes at week 176 of minus 12.3%, minus 18.7%, and minus 19.7% for the 5 mg, 10 mg, and 15 mg tirzepatide groups, compared with minus 1.3% for placebo.^[2]

The new analysis narrowed that population to 690 people assigned tirzepatide who received at least 75% of planned doses and had lost at least 5% of baseline weight by their nadir. Everyone in this analysis had prediabetes when randomized. Participants averaged 48.5 years of age, 64.6% were women, and 74.2% were White.^[1] These selection rules matter because the findings do not describe all people who start tirzepatide, those who stop treatment, or people without an initial 5% reduction.

The doses in the trial are study details, not instructions. Prescribing, dose changes, and decisions about continuing treatment require a qualified clinician who can consider indication, contraindications, adverse effects, and local product labeling.

How the researchers defined regain

Each participant's weight nadir was the lowest on-treatment weight recorded during the study. Regain was calculated as a percentage of pretreatment body weight, not as a percentage of the kilograms previously lost. The groups were less than 5%, 5% to less than 10%, and at least 10% regain from nadir by week 176.^[1]

That denominator can change how a number feels. In the three groups, median regain was 1.2 kg, 7.2 kg, and 13.7 kg. The median share of maximum weight reduction still maintained at week 176 was 94.8%, 69.9%, and 44.3%, respectively.^[1] A reader should therefore avoid comparing these categories with reports that define regain as a share of weight lost.

Weight nadir also arrived at different times. Mean time to nadir was about 21 to 23 months across dose groups, and more than half of participants in the 10 mg and 15 mg groups reached it after week 72.^[1] One weigh-in near an apparent plateau cannot identify a final long-term nadir.

Metabolic measures did not move in lockstep

At week 176, normoglycemia was recorded in 96.1% of participants with less than 5% regain, 92.6% with 5% to less than 10% regain, and 91.2% with at least 10% regain. Blood lipids and HbA1c remained improved from baseline across the regain groups, although the paper describes some attenuation in blood-pressure and waist-to-height-ratio improvements among people with the greatest regain.^[1]

Central adiposity showed a clearer gradient. From nadir to week 176, a worse waist-to-height-ratio category occurred in 6.4%, 24.3%, and 36.8% of the three regain groups. At study end, the proportion in the paper's high waist-to-height category was 45%, 59%, and 68%, respectively.^[1] These are risk markers rather than counts of heart attacks, strokes, or deaths.

The result is more nuanced than either 'the scale does not matter' or 'any regain means treatment failed.' In this selected group, greater regain tracked with more loss of some earlier improvements, yet most measured values remained better than baseline. The analysis cannot determine an individual's future clinical outcome from a regain category.

Continued treatment is part of the finding

The analysis covered participants who remained reasonably adherent during ongoing treatment. It should not be used to predict what happens after stopping tirzepatide. The main SURMOUNT-1 report separately found that 1.3% of tirzepatide participants and 13.3% of placebo participants developed type 2 diabetes by week 176. After a further 17 weeks off treatment or placebo, those figures were 2.4% and 13.7%.^[2]

Our tirzepatide evidence overview places these longer-term findings alongside other trial results. The tirzepatide safety section covers adverse effects and the limits of applying group averages to one person. Neither page replaces individual medical care.

The main randomized report found gastrointestinal events were the most common adverse events apart from COVID-19. Most were mild to moderate and occurred mainly during dose escalation.^[2] The post hoc regain paper was not designed to re-establish the full safety profile.

Limits that change the interpretation

Post hoc grouping can reveal patterns, but it cannot preserve all the protections of the original randomization. There was no placebo comparison in the regain analysis, and the groups were defined by an outcome that occurred after treatment began. The group with at least 10% regain contained only 57 people, so its estimates are less stable than those from the 484-person group with less than 5% regain.^[1]

The authors also used last observation carried forward when a week-176 weight was missing. Waist-to-height ratio was available at baseline, nadir, and week 176, but other metabolic measures were compared at baseline, week 72, and week 176. Because about 55% reached nadir after week 72, those measurements do not line up perfectly with the regain period.^[1]

Eli Lilly funded both reports. Several authors of the new analysis were company employees and shareholders, and the funder participated in design, analysis, interpretation, manuscript preparation, and the submission decision.^[1] Funding does not invalidate a result, but the declared role belongs in the interpretation.

What we do not know yet

This analysis does not establish the best response to weight regain, a dose-adjustment rule, or a threshold that predicts a clinical event. It does not tell us whether the same pattern holds for people without prediabetes, for those who receive fewer than 75% of doses, or for those who lose less than 5% before nadir. It also cannot separate drug effects from other differences between the regain groups.

The grounded takeaway is narrower. Among selected tirzepatide-treated SURMOUNT-1 participants with prediabetes who responded and stayed on treatment, limited regain after the lowest weight was common, and many metabolic improvements persisted through three years. Greater regain corresponded with weaker maintenance of some risk markers. A clinician should interpret an individual weight change alongside waist measures, glucose, blood pressure, lipids, treatment tolerance, and the person's broader health goals.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.