So how much protein should I eat on a GLP-1?

This page does not answer that, on purpose, and no regulator answers it either. What can be reported is what four named bodies have published. The joint ACLM/ASN/OMA/TOS advisory says 1.2–1.6 g/kg a day "have also been proposed during active weight reduction" — and keeps the number out of its own recommendations table. A sponsor-convened international Delphi panel recommends 1.2–1.5 g/kg of actual body weight during the weight-loss phase and grades that as expert opinion — 87% of panellists scored it 7–9; median score 7. The general adult RDA is 0.8 g/kg. At 90 kg those are 108–144 g, 108–135 g and 72 g a day. Which one applies to you is a conversation with a clinician or a registered dietitian who knows your kidney function, your calorie intake and your training.

How much muscle do you actually lose on semaglutide or tirzepatide?

The best current figure comes from a 2026 systematic review in Annals of Internal Medicine covering 35 randomised trials, median duration 26 weeks. Within the incretin groups, the median share of total weight loss attributable to reductions in muscle-based indices was 28.3%, with an interquartile range of 15.9% to 39.9%. Sixty-five per cent of studies exceeded the review's prespecified 25% benchmark. Two caveats from the review itself: DXA and bioimpedance lean mass are not the same thing as skeletal muscle, and no study in it reported any objective measure of physical function. In the lifestyle and placebo comparator arms that lost weight, 38% also exceeded the benchmark — so this is substantially what weight loss does, at a faster rate.

Is there any trial showing that eating more protein preserves muscle on a GLP-1?

No. A 2026 review states it plainly: "However, direct randomized evidence on protein supplementation during GLP-1 therapy remains absent; a registered trial is expected to provide initial data by 2027". Every gram-per-kilogram figure in circulation is extrapolated from protein trials in non-GLP-1 weight loss, which is a reasonable thing for clinicians to do and is not the same as evidence in this population. The four-society advisory is also explicit that protein is not the lever people think it is on its own: "Importantly, clinicians should understand—and emphasize to individuals taking GLP-1s—that increased protein intake alone is likely inadequate to support the preservation of muscle mass in the absence of structured resistance/strength training."

Why do different sources give different protein numbers?

Because they are different kinds of document answering slightly different questions. The 0.8 g/kg RDA is the amount that meets the requirement of 97 to 98 per cent of healthy adults — a floor for adequacy, not an optimum, and not derived in people losing 20% of their body weight. The higher figures are proposals extrapolated from protein-and-weight-loss literature. They also disagree about the denominator: the advisory notes it is "unclear whether these goals should be based on actual body weight, corrected (adjusted or ideal) body weight, or fat-free mass, as the use of actual weight can significantly overestimate protein requirements". Across every weight-based figure published, the spread at 90 kg is 72 to 144 grams a day.

What is the RDA for protein, and is 50 g a day the same thing?

They are two different official numbers for two different purposes. The Institute of Medicine's Recommended Dietary Allowance is 0.8 g/kg of body weight per day for adults, which the DRI tables express as 56 g a day for men and 46 g for women, within an Acceptable Macronutrient Distribution Range of 10 to 35 per cent of energy. The 50 g figure is the FDA's Daily Reference Value, set in 21 CFR 101.9 for percentage labelling on the Nutrition Facts panel and "based on the reference caloric intake of 2,000 calories" — it is a labelling constant, not a personal requirement, and it is not adjusted for body weight at all.

Does this page tell me my protein target?

No, and it will not. The calculator above multiplies your body weight by each published figure and shows you four separate answers, each labelled with who published it, how they graded it and who paid for the panel. It does not average them, does not highlight one and does not add a fifth. At 90 kg the published figures run from 72 to 144 grams a day, none of them is from a regulator, the most authoritative one is not in its own recommendations table, and no randomised trial has tested any of them in a person on a GLP-1. A number from us would be the worst-sourced entry in that list.

How Much Protein on a GLP-1? Every Published Figure, Sourced

Who has actually published a protein figure, and what it is worth

Both drug labels state that the weight comes off as more fat than lean tissue, and neither puts a number on the lean part. WEGOVY’s section 12.2 reads, in full on this point: "Semaglutide lowers body weight with greater fat mass loss than lean mass loss." ZEPBOUND’s carries the same sentence with tirzepatide in it. That is the whole of what the FDA-approved labelling says about body composition, and neither document mentions dietary protein at all — the word does not occur anywhere in the WEGOVY label, and its single appearance in ZEPBOUND’s is C-reactive protein in a blood-pressure result.

The magnitude comes from elsewhere, and it is real. A 2026 systematic review in Annals of Internal Medicine pooled 35 randomised trials of liraglutide, semaglutide, tirzepatide and dulaglutide and found that "Within the incretin groups across agents and measurement methods, the median proportion of total weight loss attributable to reductions in muscle-based indices was 28.3% (IQR, 15.9% to 39.9%), with 65% exceeding the benchmark of about 25%". That is the number worth knowing. The same review found the same thing in the comparator arms: lean-mass loss exceeded the benchmark in 38% of the lifestyle and placebo groups that lost weight too. Losing lean tissue while losing weight is what losing weight does; the drugs do it faster because they take off more weight.

Then comes the part everybody quotes and nobody sources. Four bodies have published a protein figure for people on these drugs. The joint advisory of the American College of Lifestyle Medicine, the American Society for Nutrition, the Obesity Medicine Association and The Obesity Society writes that "Higher targets, such as 1.2–1.6 g/kg body weight/day [155], have also been proposed during active weight reduction." — note the passive, and note that its own Table 6 of key dietary recommendations carries no gram figure at all. The Delphi panel is more direct: "During the weight-loss phase on GBT, a protein intake of 1.2–1.5 g/kg actual body weight/day or equivalent to 25–30 % energy on a 1600 kcal/day diet, is recommended." The panel grades that statement, itself, as expert opinion; 87% of panellists scored it 7–9; median score 7.

None of which is tested. The most recent review of the field states plainly that "However, direct randomized evidence on protein supplementation during GLP-1 therapy remains absent; a registered trial is expected to provide initial data by 2027". So the honest position is that a reasonable, widely-shared, entirely untested range exists, that the general adult RDA of 0.8 g/kg a day sits well below it, and that the four-society advisory's own warning is the sentence most worth carrying away: "Importantly, clinicians should understand—and emphasize to individuals taking GLP-1s—that increased protein intake alone is likely inadequate to support the preservation of muscle mass in the absence of structured resistance/strength training."

This page states no protein figure of its own, and there is no cell anywhere on it with our number in it. We are a directory. Every figure below belongs to somebody named, and is printed with their grade and their funding attached so you can weigh it yourself.

The formula

grams per day = g/kg figure × body weight (kg)

— per SOURCE, never combined —

1.2–1.6 g/kg → 108–144 g at 90 kg

1.2–1.5 g/kg → 108–135 g at 90 kg

≥ 0.8 g/kg → ≥ 72 g at 90 kg

0.8 g/kg → 72 g at 90 kg

The multiplication is trivial; which number goes into it is the entire question, and the four answers above are 72 grams a day apart at the same body weight. Nothing on this page averages them, and there is no fifth figure of ours. Two further complications the sources raise themselves: the advisory notes it is "unclear whether these goals should be based on actual body weight, corrected (adjusted or ideal) body weight, or fat-free mass, as the use of actual weight can significantly overestimate protein requirements", and a protein intake is only meaningful against a calorie intake, which a GLP-1 changes.

Worked example

A man of 90 kg (198 lb) on tirzepatide, asking each source in turn:

FDA-approved labelling: no figure. Neither label mentions dietary protein.
IOM/NASEM RDA for the general adult population: 0.8 g/kg → 72 g a day
FDA Daily Value on a Nutrition Facts label: 50 g a day, not weight-based
Four-society advisory, active weight reduction: 1.2–1.6 g/kg → 108–144 g a day — but stated as "proposed", and absent from its own recommendations table
Delphi consensus, weight-loss phase: 1.2–1.5 g/kg → 108–135 g a day — graded Expert opinion, sponsor-convened panel
Delphi consensus, maintenance: ≥ 0.8 g/kg → ≥ 72 g a day
Spread across every weight-based figure published: 72 g to 144 g a day, from 4 sources, at one body weight. No trial has tested any of them in a person taking a GLP-1.

Every published protein figure for adults on a GLP-1

One row per named source, strongest provenance first. The last two columns are the ones that are never printed anywhere else, and they are the reason the figures in column one should not be read as equivalent.

Source	What it states	Kind of statement	Grade its own authors give it	Disclosed funding
WEGOVY labelno phase stated	None — the word "protein" does not appear in the labelstated for people on a GLP-1	FDA-approved drug label	FDA-approved labelling	None disclosed
ZEPBOUND labelno phase stated	None — the only "protein" in the label is C-reactive protein, in section 14.5stated for people on a GLP-1	FDA-approved drug label	FDA-approved labelling	None disclosed
FDA Daily ValueGeneral adult population	50 g a daygeneral population	US regulation	Codified in regulation	None disclosed
IOM / NASEM RDAGeneral adult population	0.8 g/kg a day (56 g men, 46 g women)general population	Dietary Reference Intake	Recommended Dietary Allowance — "set to meet the needs of almost all (97 to 98 percent) individuals in a group"	None disclosed
ACLM / ASN / OMA / TOS advisoryDuring active weight loss	1.2–1.6 g/kg a day, "have also been proposed"stated for people on a GLP-1	Joint advisory of named societies	Not graded. Written in the passive as figures that "have been proposed", and the advisory’s own Table 6 of key dietary recommendations carries no protein number	None disclosed
Delphi consensus, maintenance phaseDuring maintenance	≥ 0.8 g/kg a daystated for people on a GLP-1	Modified-Delphi expert consensus	Expert opinion	Steering committee "appointed by the sponsor"; Nestlé Health Science funded project management and medical writing and reimbursed travel
Delphi consensus, weight-loss phaseDuring active weight loss	1.2–1.5 g/kg a day, or 25–30% of energy on 1,600 kcalstated for people on a GLP-1	Modified-Delphi expert consensus	Expert opinion	Steering committee "appointed by the sponsor"; Nestlé Health Science funded project management and medical writing and reimbursed travel

Read from the primary documents, not from each other. Two things to take from the right-hand columns. First, no regulator appears in the top half with a figure: the string "protein" does not occur anywhere in the WEGOVY label, and its single occurrence in ZEPBOUND's is C-reactive protein in section 14.5. Second, the two GLP-1-specific ranges are not the same kind of object — one is a sentence in the passive voice in a discussion section, the other is a voted statement that its own panel grades as expert opinion. The general-population rows are here because the figures above them are routinely quoted as though 0.8 g/kg were a floor nobody serious uses, and it is the RDA.

Where each figure sits inside its own document

The distinction that decides how much weight a figure carries: a number in a source’s recommendations is a recommendation, and a number in its discussion is a report of what others have proposed.

Source	Where the figure appears	In the source’s own recommendations?	Voting or consensus recorded
WEGOVY labelFDA-approved drug label	Nowhere — the string "protein" does not occur in the document	No recommendation either way	None reported
ZEPBOUND labelFDA-approved drug label	Section 14.5, and it is C-reactive protein — not dietary protein	No recommendation either way	None reported
ACLM / ASN / OMA / TOS advisoryJoint advisory of named societies	Body text of section 4.6, "Preservation of muscle and bone mass". NOT in Table 6, the advisory's own "Key dietary recommendations to support effective GLP-1 therapy", which lists only "Lean proteins (e.g., poultry, fish/seafood) and eggs" and carries no gram figure	No — discussion only	None reported
Delphi consensus, weight-loss phaseModified-Delphi expert consensus	Table 4, as a voted consensus statement	Yes	87% of panellists scored it 7–9; median score 7
Delphi consensus, maintenance phaseModified-Delphi expert consensus	Table 5, as a voted consensus statement	Yes	93% of panellists; median score 7

The middle row is the finding. The four-society advisory is the most authoritative document on this subject — four US organisations, published simultaneously in four journals — and it kept the gram figure out of its own recommendations. Its Table 6, "Key dietary recommendations to support effective GLP-1 therapy", lists "Lean proteins (e.g., poultry, fish/seafood) and eggs" under factors to encourage and carries no number anywhere. The 1.2–1.6 g/kg range lives in the body of section 4.6, in one passive sentence: "Higher targets, such as 1.2–1.6 g/kg body weight/day [155], have also been proposed during active weight reduction." That is not the same as recommending it, and the difference is worth a reader's time.

What is actually established about lean mass

The finding and the recommendation are different objects. This table is only the finding — what the labels state, what the trials measured, and what has never been tested.

Source	What it states	Is a figure given?	The figure, or why there is none
WEGOVY label, section 12.2FDA label	“Semaglutide lowers body weight with greater fat mass loss than lean mass loss.”	No	No figure. The label states the direction and stops.
ZEPBOUND label, section 12.2FDA label	“Tirzepatide lowers body weight with greater fat mass loss than lean mass loss.”	No	No figure. The same sentence, with the other molecule in it.
Annals of Internal Medicine systematic review, 2026peer-reviewed	“Within the incretin groups across agents and measurement methods, the median proportion of total weight loss attributable to reductions in muscle-based indices was 28.3% (IQR, 15.9% to 39.9%), with 65% exceeding the benchmark of about 25%”	Yes	28.3% of total weight lost, median across 35 trials (IQR 15.9% to 39.9%)
The same review, on diet and exercise alonepeer-reviewed	“the 13 studies reporting weight loss in the lifestyle or placebo comparator groups accounted for a median weight loss of -2.5% (IQR, -4.5% to -0.9%), 38% of which exceeded the respective benchmark”	Yes	Lean-mass loss above the same benchmark occurred in 38% of non-drug interventions too
Whether eating more protein changes any of itpeer-reviewed	“However, direct randomized evidence on protein supplementation during GLP-1 therapy remains absent; a registered trial is expected to provide initial data by 2027”	No	Untested. No randomised trial has reported.

Two cautions the sources raise about their own numbers. The 28.3% median is lean mass and fat-free mass measured by DXA and bioimpedance, and the 2026 review of this literature notes that "DXA-derived lean mass and BIA-derived fat-free mass are not equivalent to skeletal muscle" — those methods count water, organs and connective tissue too. And the same review's conclusion is not that this is settled: "Future trials are needed to better understand mechanisms underlying these changes and their clinical implications." Not one study in it reported an objective measure of physical function.

Frequently asked questions

So how much protein should I eat on a GLP-1?: This page does not answer that, on purpose, and no regulator answers it either. What can be reported is what four named bodies have published. The joint ACLM/ASN/OMA/TOS advisory says 1.2–1.6 g/kg a day "have also been proposed during active weight reduction" — and keeps the number out of its own recommendations table. A sponsor-convened international Delphi panel recommends 1.2–1.5 g/kg of actual body weight during the weight-loss phase and grades that as expert opinion — 87% of panellists scored it 7–9; median score 7. The general adult RDA is 0.8 g/kg. At 90 kg those are 108–144 g, 108–135 g and 72 g a day. Which one applies to you is a conversation with a clinician or a registered dietitian who knows your kidney function, your calorie intake and your training.
Do the Wegovy and Zepbound labels say anything about protein?: No. The word "protein" does not occur anywhere in the WEGOVY label, and it occurs once in ZEPBOUND's — in section 14.5, where it is high-sensitivity C-reactive protein in a sleep-apnoea result. What both labels do state, in section 12.2, is the direction of the body-composition change: "Semaglutide lowers body weight with greater fat mass loss than lean mass loss." and the identical sentence for tirzepatide. No magnitude, no protein guidance. The only quantified diet instruction in either document is caloric — ZEPBOUND's section 14.1 describes a lead-in with "approximately 1,200 kcal/day (females) or 1,500 kcal/day (males)".
How much muscle do you actually lose on semaglutide or tirzepatide?: The best current figure comes from a 2026 systematic review in Annals of Internal Medicine covering 35 randomised trials, median duration 26 weeks. Within the incretin groups, the median share of total weight loss attributable to reductions in muscle-based indices was 28.3%, with an interquartile range of 15.9% to 39.9%. Sixty-five per cent of studies exceeded the review's prespecified 25% benchmark. Two caveats from the review itself: DXA and bioimpedance lean mass are not the same thing as skeletal muscle, and no study in it reported any objective measure of physical function. In the lifestyle and placebo comparator arms that lost weight, 38% also exceeded the benchmark — so this is substantially what weight loss does, at a faster rate.
Is there any trial showing that eating more protein preserves muscle on a GLP-1?: No. A 2026 review states it plainly: "However, direct randomized evidence on protein supplementation during GLP-1 therapy remains absent; a registered trial is expected to provide initial data by 2027". Every gram-per-kilogram figure in circulation is extrapolated from protein trials in non-GLP-1 weight loss, which is a reasonable thing for clinicians to do and is not the same as evidence in this population. The four-society advisory is also explicit that protein is not the lever people think it is on its own: "Importantly, clinicians should understand—and emphasize to individuals taking GLP-1s—that increased protein intake alone is likely inadequate to support the preservation of muscle mass in the absence of structured resistance/strength training."
Why do different sources give different protein numbers?: Because they are different kinds of document answering slightly different questions. The 0.8 g/kg RDA is the amount that meets the requirement of 97 to 98 per cent of healthy adults — a floor for adequacy, not an optimum, and not derived in people losing 20% of their body weight. The higher figures are proposals extrapolated from protein-and-weight-loss literature. They also disagree about the denominator: the advisory notes it is "unclear whether these goals should be based on actual body weight, corrected (adjusted or ideal) body weight, or fat-free mass, as the use of actual weight can significantly overestimate protein requirements". Across every weight-based figure published, the spread at 90 kg is 72 to 144 grams a day.
What is the RDA for protein, and is 50 g a day the same thing?: They are two different official numbers for two different purposes. The Institute of Medicine's Recommended Dietary Allowance is 0.8 g/kg of body weight per day for adults, which the DRI tables express as 56 g a day for men and 46 g for women, within an Acceptable Macronutrient Distribution Range of 10 to 35 per cent of energy. The 50 g figure is the FDA's Daily Reference Value, set in 21 CFR 101.9 for percentage labelling on the Nutrition Facts panel and "based on the reference caloric intake of 2,000 calories" — it is a labelling constant, not a personal requirement, and it is not adjusted for body weight at all.
Does this page tell me my protein target?: No, and it will not. The calculator above multiplies your body weight by each published figure and shows you four separate answers, each labelled with who published it, how they graded it and who paid for the panel. It does not average them, does not highlight one and does not add a fifth. At 90 kg the published figures run from 72 to 144 grams a day, none of them is from a regulator, the most authoritative one is not in its own recommendations table, and no randomised trial has tested any of them in a person on a GLP-1. A number from us would be the worst-sourced entry in that list.

Sources

Embed this calculator

Free to use on your own site, no permission needed and no account. The widget resizes itself to fit your layout. Keeping the credit line underneath is appreciated, but the calculator works exactly the same if you delete it.

Paste this

<!-- Protein Intake on a GLP-1 — by Alpha Health Finder. Free to embed; keeping the credit line below is appreciated but not required. -->
<iframe src="https://alphahealthfinder.com/embed/tools/glp1-protein-target"
        title="Protein Intake on a GLP-1"
        width="100%" height="2780" loading="lazy"
        style="border:1px solid #e2e8f0;border-radius:12px;max-width:680px;width:100%"
        data-ahf-embed="glp1-protein-target"></iframe>
<p style="max-width:680px;margin:8px 0 0;font:13px/1.5 system-ui,-apple-system,sans-serif;color:#64748b">
  <a href="https://alphahealthfinder.com/tools/glp1-protein-target">Protein Intake on a GLP-1</a> by <a href="https://alphahealthfinder.com">Alpha Health Finder</a>
</p>
<script src="https://alphahealthfinder.com/embed/resize.js" async></script>

Next steps

Providers that include a dietitian, not just a penA protein figure is only worth having if someone is looking at your labs, your calories and your training alongside it. Compare what each platform actually includes beyond the prescription.What the GLP-1 trials measured, and what they did notThe trial results, the body-composition findings and the questions the systematic reviews say are still open.