Tirzepatide 15 mg: 1.60 kg of Fat-Free Mass Lost

Obesity Science and Practice published a systematic review and meta-analysis on 6 September 2026 by Ampofo, Apprey, Amoako and Turkson, from Kwame Nkrumah University of Science and Technology in Kumasi, Ghana.

It pools 19 randomised trials from the SURMOUNT, STEP, SCALE and OASIS programmes, and it asks a question those trials were not designed to answer. Not how much weight came off, but what came off with it.

A framing note. These are approved medicines and this is a synthesis of clinical trial data. Peptra Labs supplies Tirzepatide as a reference material for laboratory research only, and nothing below is guidance about human use.

What the tirzepatide analysis measured

The methods are the conventional ones and the authors state them. PRISMA 2020 for reporting, the Cochrane RoB2 tool for risk of bias, GRADE for certainty of evidence, and continuous outcomes pooled as mean differences in a random-effects model.

Daily energy intake declined by 24.00 to 39.20 percent across drug classes, with model-estimated daily deficits reaching 1200 kcal.

The compound-specific figure is the one to hold onto. Tirzepatide 15 mg was associated with a mean fat-free mass reduction of 1.60 kg, representing 2.80 percent of body weight.

Fat-free mass is everything that is not adipose tissue, which includes muscle, organ mass and body water. Losing 2.80 percent of body weight as fat-free mass is not the same measurement as losing 2.80 percent of fat-free mass, and the abstract reports the former.

Where the tirzepatide reporting and the laboratory disagree

The pair of numbers that makes this paper worth reading sits in the results.

Investigator-reported malnutrition occurred in 0.12 percent of participants. Objective laboratory screening identified total lymphocyte counts below 910 per microlitre in 2.90 percent of participants on active therapy, against 1.77 percent in placebo arms. The authors conclude that standard adverse event reporting underestimates true nutritional risk.

Those two figures are produced by two different processes. An investigator-reported adverse event requires somebody to notice a problem, decide it is an event, and code it under a term. A laboratory value is measured on everyone the protocol says to measure, whether or not anyone was looking for it.

That is the whole argument, and it is a good one. It is also worth reading precisely.

The difference between 2.90 and 1.77 percent is 1.13 percentage points, and the abstract gives no confidence interval for that comparison. A single lymphocyte threshold is a screening flag rather than a diagnosis of malnutrition, and a background rate exists in the placebo arms. The honest reading is that objective screening finds a signal at a rate the case report forms did not capture, not that roughly one in thirty-five participants became malnourished.

The figure that is not about tirzepatide

One number in the results belongs to a different compound and the article should say so.

Pancreatic lipase increased by a mean of 31 percent in the SCALE programme, which the authors describe as a secondary metabolic signal. SCALE is the liraglutide programme, so that figure is not a tirzepatide result. The energy intake range is likewise given across drug classes rather than per agent.

Of the headline numbers, only the 1.60 kg is specific to tirzepatide. Anyone quoting this paper should carry that distinction with them, because a pooled analysis of four programmes reads as though every figure applies to every agent, and it does not.

What the authors propose next

The conclusion moves from finding to recommendation, which is a step past what a meta-analysis can establish on its own.

The authors propose a Tiered Stepped-Care Algorithm mandating baseline screening of albumin and total lymphocyte count, periodic monitoring at weeks 12, 24 and 52, and defined intervention thresholds, aimed at preventing sarcopenia and frailty, particularly in adults aged 65 and over.

That is a proposal built on the evidence they synthesised, not a protocol that has been tested. No trial in this analysis randomised participants to the algorithm, and the paper offers no data on whether the monitoring schedule changes any outcome. It is a reasonable idea awaiting its own study, and the authors present it as a proposal rather than a finding.

The disclosure on the tirzepatide analysis

The competing interests statement is short and it says something useful.

The authors declare no direct financial conflicts concerning the review. They then note that the 19 analysed trials were primarily industry-funded by Eli Lilly or Novo Nordisk, and that this independent synthesis received no corporate support.

That combination is the interesting part. An unfunded academic group re-analysing industry trial data for a harm those trials reported at 0.12 percent is structurally different from a manufacturer-adjacent summary of the same literature. Yesterday we covered a narrative review that put mean weight loss at 15 to 20 percent and carried a disclosure of paid advisory relationships with both of those companies. Neither paper is dishonest. They are simply looking at the same trials from different positions, and the reader gets a fuller picture holding both.

The limits of a trial-level synthesis

This is a meta-analysis of published trial-level data rather than individual participant data, which constrains what it can resolve.

Body composition can be measured by dual-energy X-ray absorptiometry or by bioimpedance, and the abstract does not state which methods the pooled trials used, so the variation those choices introduce is not visible in a pooled mean difference. The lymphocyte comparison is pooled across four programmes with different populations, comparators and durations. The abstract states that GRADE was applied without reporting the resulting certainty ratings, so the strength the authors assign to their own findings is not visible from the abstract alone.

We have written about a systematic review that could not pool its numbers at all for exactly this class of reason, and about a post hoc analysis of composite endpoints from the trial programme behind tirzepatide.

What it means for the laboratory

The transferable point is about which endpoint gets measured, not about clinical care.

Weight is the cheapest outcome to record and the least informative about composition. This analysis is an argument that a study of any compound in the tirzepatide class which reports only mass has left the more interesting half of the result unmeasured. Retatrutide adds a third receptor to the same pharmacology, and the same question applies to it with less published data behind it.

The other half of the picture is duration. We covered an estimate of how fast weight returns after stopping that put monthly regain at 1.04 kg, and regained weight does not necessarily return in the same tissue that was lost.

For laboratories, our European research buyer guide covers procurement and documentation, and everything is supplied on a research use only basis.

References

• Ampofo E, Apprey C, Amoako M, Turkson FD. A Systematic Review and Meta-Analysis of Malnutrition and Metabolic Failure in High-Potency Incretin Therapy. Obesity Science and Practice, 6 September 2026, 12(5):e70188. doi 10.1002/osp4.70188
• Gill J, Sumithran P. The Evolution of Medications for Obesity Management. JGH Open, 4 September 2026, 10(9):e70466. doi 10.1002/jgh3.70466
• Kow CS, Thiruchelvam K, Ramachandram DS, Zaihan AF. Weight Regain Trajectories After Discontinuation of Semaglutide or Tirzepatide. Endocrinology, Diabetes and Metabolism, September 2026, 9(5):e70325

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.