Tirzepatide Regain: 1.04 kg a Month, 15 Months Modelled
A Bayesian re-analysis published in Endocrinology, Diabetes and Metabolism in September 2026 estimates what happens to body weight after people stop taking tirzepatide or semaglutide.
The headline figures are 1.04 kg regained per month, half the lost weight back at 7.5 months, and a return to starting weight at 15 months.
The first of those was measured. The last one was calculated by assuming the first continues in a straight line for longer than anybody watched.
A framing note. These are approved medicines and this is an analysis of trials in patients. Peptra Labs supplies Tirzepatide as a reference material for laboratory research only, and nothing below is guidance about human use, about weight loss, or about stopping any treatment.
How the tirzepatide estimate was built
This is not a new trial. The authors took published timepoint-level aggregate data from studies that reported weight change after discontinuation, and re-analysed it with a Bayesian hierarchical longitudinal model.
Six tirzepatide and semaglutide studies met the criteria, comprising ten intervention arms and 1,776 participants. Observed follow-up ranged from 4 to 52 weeks.
The model was asked to estimate four things: weight loss at the moment of stopping, the monthly rate of regain, the time to regain half of what was lost, and the time to return to baseline weight. A Bayesian meta-regression then examined three possible predictors of faster regain: which medication, how much weight had been lost initially, and whether behavioural or lifestyle support continued afterwards.
Reconstructing individual trajectories from aggregate published figures is a legitimate technique and a demanding one. It also means every number below inherits whatever the source papers reported, including their follow-up limits.
What the tirzepatide model produced
| Estimate | Value | 95 percent credible interval |
|---|---|---|
| Weight loss at cessation | 15.35 kg | 11.18 to 19.60 |
| Monthly regain | 1.04 kg | 0.80 to 1.29 |
| Time to regain half | 7.50 months | 5.05 to 10.57 |
| Time to return to baseline | 15.00 months | 10.10 to 21.13 |
On the three predictors, the authors are careful. Tirzepatide showed numerically faster regain than semaglutide in unadjusted analyses, but no clear independent drug-specific difference remained after adjustment. Greater initial weight loss showed the strongest directional association with faster regain, and the authors state that its credible interval included zero. Behavioural or lifestyle support was directionally associated with slower regain, with the estimate described as imprecise.
Three signals, all pointing somewhere, none of them established. That is an unusually honest set of sentences for this topic.
Where the tirzepatide data stop and the line continues
Observed follow-up ran from 4 to 52 weeks. The return-to-baseline estimate is 15 months, which is about 65 weeks.
The methods state that longer-term estimates assumed a constant linear regain rate, and the conclusion repeats it: estimates beyond 52 weeks were model-based extrapolations and should be interpreted cautiously.
So the 15-month figure is not something anybody observed. It is what happens if you take a measured rate and continue it past the end of the measurements.
Whether that assumption holds is an open biological question. Regain could plateau as appetite regulation resettles, it could accelerate, or it could vary by how much was lost. The model cannot distinguish those, because after week 52 there is nothing left to fit.
None of this makes the number wrong. It makes it a projection, and the authors say so. The distinction survives only if whoever repeats the figure repeats the caveat with it, which in practice almost never happens.
What the result does establish
Strip out the extrapolation and a solid finding remains.
Within the observed window, regain after stopping tirzepatide or semaglutide was rapid and substantial. A rate of 1.04 kg per month with a credible interval from 0.80 to 1.29 is a tight estimate, and 7.5 months to lose half the benefit sits at the edge of the observed range rather than beyond it.
The absence of a clear drug-specific difference after adjustment is also worth keeping. The unadjusted signal favouring faster regain on tirzepatide did not survive, which is what you would expect if the driver is how much weight came off rather than which molecule took it off.
That reading is consistent with the directional finding on initial weight loss, although the authors are explicit that its interval included zero.
Where it sits in a month of reading
We have spent August on studies where weight change and outcomes come apart.
In a matched cohort where hard liver outcomes came out level, one arm dropped more body mass index and the hard endpoints did not separate. In a population with a BMI under 27, a signal appeared in people with almost nothing to lose. And last week surgery removed far more liver fat and tied on stiffness against a GLP-1 agonist.
This paper is the mirror image and it is about the weight itself rather than about an organ. Weight is the endpoint that responds fastest, moves furthest and reverses soonest. Everything downstream of it moves on slower timescales that none of these studies has followed.
Read alongside the systematic review published today in Annals of Internal Medicine, which puts placebo-subtracted weight loss at 19.0 percent for tirzepatide and 22.1 percent for Retatrutide, the two papers describe opposite ends of the same curve. One measures how far it goes down. The other estimates how fast it comes back.
What this means for the catalogue
Nothing in this paper is a laboratory result, and the transferable content is about how a number gets extended.
Model-based extrapolation beyond observed data is standard practice and it is fine when it is labelled. The failure mode is that the label falls off in transit. A preclinical example is the same shape: a dose-response curve fitted across four concentrations, then an effect quoted at a fifth that was never tested.
The habit is to ask where the data stopped before asking what the number says.
For laboratories, our European research buyer guide covers procurement and documentation, everything is supplied on a research use only basis, and our ranking of the most-studied research peptides counts primary studies for each compound.
References
- Kow CS, Thiruchelvam K, Ramachandram DS, Zaihan AF. Weight Regain Trajectories After Discontinuation of Semaglutide or Tirzepatide: A Reconstructed Aggregate-Data Bayesian Longitudinal Meta-Analysis. Endocrinology, Diabetes and Metabolism, September 2026, 9(5):e70325. doi 10.1002/edm2.70325
- Moiz A, Filion KB, Samuels AE, Tsoukas MA, Yu OHY, Peters TM, Eisenberg MJ. Efficacy and Safety of Glucagon-like Peptide-1 Receptor Agonists and Co-agonists for Weight Loss Among Adults Without Diabetes: An Updated Systematic Review. Annals of Internal Medicine, 1 September 2026. doi 10.7326/ANNALS-25-05519
- Nicholls SJ, Pavo I, Bhatt DL, et al. Cardiovascular Outcomes with Tirzepatide versus Dulaglutide in Type 2 Diabetes. New England Journal of Medicine, 18 December 2025, 393(24):2409-2420
The products referenced on this site are supplied for laboratory research use only. They are not medicines and are not intended for human or veterinary use. This article summarises published research for informational purposes and is not medical advice. Statements about third-party studies belong to their authors.
