The Plateau Is the Drug Working
Retatrutide is an investigational research peptide with an unfinished phase 3 program. Every retatrutide user meets the plateau in time, and the usual conclusion is that the compound has stopped working. The trial data point the other way, because the plateau is the body defending its weight and the maintenance dose is how you live with that defense. This article is community-sourced biology education for research-use contexts, and nothing in it is medical advice.
Retatrutide is an investigational research peptide with an unfinished phase 3 program. Every retatrutide user meets the plateau in time, and the usual conclusion is that the compound has stopped working. The trial data point the other way, because the plateau is the body defending its weight and the maintenance dose is how you live with that defense. This article is community-sourced biology education for research-use contexts, and nothing in it is medical advice.
I · The body defends its weight
The plateau is the body defending the weight it now holds, and that defense is a hormonal response, which means the drug is doing exactly what it should.
The weight-loss curve that looks like a flatline is a stalemate between the drug and the system it works on, because every pound of loss triggers a set of hormonal adjustments that push appetite up and energy output down. That response is ancient, and it runs through pathways that predate any diet, since the gut hormones and adipose signals feed the brain centers that read them, and the whole loop acts as one coordinated defense of the weight the body has carried. The community’s most experienced protocol voices describe this as the body fighting in both directions, and the same defense that resists loss also resists gain, which is why regain after stopping is as predictable as the plateau itself 8.
The measurements behind that description come from metabolic wards, where intake and output are weighed rather than estimated. Kevin Hall’s energy-balance work, which the community corpus cites over and over, shows that most of the plateau comes from appetite returning, so people eat more without noticing, with a smaller contribution from the drop in energy expenditure that accompanies any weight loss 4. A 2025 metabolic-ward study of tirzepatide cited to Ravussin and colleagues in Cell Metabolism found the treated group lost about 38.5 lbs versus 18.5 lbs on placebo, while total daily energy expenditure fell by about 414 kcal versus 197, and sleeping metabolic rate dropped about 204 kcal versus 94 3. Those numbers are the counter-response measured in the ward: the drug pulls intake down, and the body pushes its own levers back up.
Thyroid output follows the same pattern, since a calorie deficit suppresses thyroid hormone production, which is a normal adaptation rather than a disorder 8. Community members who run labs through a long loss see low T3 and falling IGF-1 for most of the way down, with SHBG rising in parallel, and all of it corrects at maintenance, so the honest reading of a plateau is that the endocrine system is doing what it evolved to do 8. The community line that sums up the whole section: “the body doesn’t think it needs to lose weight anymore… you’ll be at a maintenance dose and just feel normal” [anecdotal] 8.
The plateau is the system working as designed, so a stall is a signal about the dose and the timeline, and the stall that means the drug stopped working is the rare one.
The practical math of a plateau is that intake and output have met again, so the deficit that drove the first twenty pounds has quietly closed. Kevin Hall’s modeling shows the closing happens mostly through the fork, which is why the lever at a plateau is the dose and the deficit plan, and why the corpus is blunt about the last explanation to reach for: eating too little 4.
II · The receptor keeps answering
The receptor is still answering, because the evidence for receptor burnout is thin and the evidence for a defended weight is strong.
Tolerance is the first word most people reach for at a plateau, so it helps to be precise about what tolerance would require: a receptor that stops responding to its ligand, leaving the drug without a grip at any dose. The receptor-reset story, which says a break will restore sensitivity, is the myth the community argues about most, and one of its most experienced voices is categorical about it: “That’s a popular lay theory but there’s zero evidence to support it and a large amount of evidence against it” [anecdotal] 8.
The biology of receptor handling points the same way, because GLP-1 receptors internalize and recycle on a timescale of minutes to hours, reaching a new steady state within hours of the first dose rather than over the weeks a reset would need 5. The drug itself accumulates over weeks, so a stall in week two is most often a system still building up, since the community’s steady-state math shows a new user on 1mg twice still carrying only a fraction of a full maintenance load, and the standard explanation to a week-one non-responder is that the actual concentration on board is equivalent to about 2mg until the drug has had time to stack 8.
The strongest evidence that the receptor keeps answering is what happens when the drug leaves, because regain is the default: “Most of the folks who stop taking GLP-1s regain most of their weight in a couple years” [anecdotal] 8. A receptor that had burned out would make stopping a non-event, whereas a body that regains is a body confirming that the drug was still holding the line.
The trial timing confirms the pattern, since in SURMOUNT-1, the tirzepatide registration trial led by Dr. Ania Jastreboff, about half of participants had reached their plateau by weeks 24 to 36, close to 90% by week 72, and the extension followed people out to 176 weeks 2. The retatrutide phase 2 trial in the New England Journal of Medicine ended while its highest-dose group was still losing weight, which is why some community members read it as a drug with no plateau, whereas the phase 3 data show the usual flattening 1. The community’s reading of the long record is that people plateau about eighteen months into therapy and then hold that weight for years without the drug losing its grip 8.
A plateau that starts between month six and month nine of therapy is the statistical norm, and a plateau that holds for months is expected rather than alarming. The question at that point is maintenance, and the drug’s continuing effect is what makes the maintenance hold 2.
Nearly 90% of SURMOUNT-1 participants had stopped losing by week 72, with about half reaching their plateau between weeks 24 and 36, so a long stall is the typical trajectory rather than a complication 2. The retatrutide phase 2 trial ran out of runway before its plateau, which is why the community reads “no plateau” in the phase 2 curve as a trial-length artifact 1.
III · Each dose carries its own plateau
Each dose carries its own plateau weight, which is why choosing a dose is choosing where the scale will settle.
The plateau lands at a different weight for every dose, because a higher dose holds appetite suppression harder, which pushes the defended weight lower, and the community states the principle in a single line: “the higher the dose the deeper the plateau” [anecdotal] 8. The corollary follows in turn, since “All doses eventually become maintenance doses once you’ve lost enough weight” [anecdotal] 8, which means a stall is a statement about the current dose more than it is a statement about the drug.
The dose-response data give the principle its numbers, because the retatrutide phase 2 trial showed average losses of 8.7% at 1mg and 24.2% at 12mg at 48 weeks 1. The step from the 2-4-8 escalation to the 2-4-8-12 escalation moved the average from 21.7% to 24.2% 7. The per-milligram return shrinks as the dose grows, so total effect keeps scaling to the top of the tested range while each added milligram buys less, which reconciles the experience of the people who respond at 1mg with the majority who need the upper range 8.
The same principle explains the most confusing stalls in the community, which are the switches from high-dose tirzepatide. Someone coming off 15mg of tirz and starting reta at 2mg feels like the drug is dead, because the new dose is a fraction of their previous exposure, and the community’s consistent advice is to expect nothing until the 6-9mg range 8. A stall that early in a switch is an under-dose wearing the costume of a plateau, and the fix is escalation, which is why the protocol voices push stalled switchers up the ladder rather than down 8.
Phase 2 averages at 48 weeks were 8.7% at 1mg and 24.2% at 12mg, so the first milligrams carry the most effect and each doubling buys less 1. The community doctrine that follows is that the total dose reached matters more than the speed of getting there, since the phase 2 fast-escalation group that moved 4mg to 8mg was still losing weight at 44 weeks 7.
IV · Tuning the dose to the goal
Maintenance is dose-finding in the other direction, because you lower the dose until the scale stops moving and that dose becomes your maintenance dose.
The maintenance dose is discovered rather than prescribed, since it is the dose at which you can eat maintenance calories with ease and hold the weight you want, and the community states it without hedging.
“Your maintenance dose is a very individual thing. It’s basically just whatever dose will keep you at the weight you want to be at.”
Community member, Retatrutide protocol corpus, 2026
The joke version of the principle is the account’s illustrative math, where “at 7.216mg you neither gain nor lose weight. That’s your maintenance dose” [anecdotal] 8. The serious version is the plateau itself, since “If you’ve plateaued then you’re currently at your maintenance dose” [anecdotal] 8, which reframes the stall from a problem into a data point: the dose you are on is holding a weight, and the only question is whether that weight is the goal.
Because the body defends in both directions, the dose works as a dial rather than an on-switch, and the community turns that dial on purpose, running the dose below maintenance to allow a bulk and above it to cut, a trick one protocol voice calls a bodybuilding cheat code [anecdotal] 8. The other boundary is excess, since “Too much ain’t any better than too little” [anecdotal] 8, and a dose that causes side effects or erases appetite is a dose that needs to come back down.
The community’s reported titration practice is to hold four weeks at a dose before judging it, since the drug accumulates, describing escalation when the effect fades and the dose is tolerated and holding when side effects appear 8. The community’s guardrail is the trial schedule itself: “I wouldn’t increase my dose faster than they did in the clinical trials” [anecdotal] 8. And for people stalled with room left on the ladder, the push is to move: “You don’t even need to waste a month with no weight loss. If it doesn’t seem to be working as well as you’d want and you’re not having issues with side effects, go up” [anecdotal] 8.
The community’s judging window is four or more weeks at a dose; its described practice is to escalate when the effect fades and the dose is tolerated, hold when side effects appear, and stay at or below the phase 3 ladder of 2 to 4 to 6 to 9 to 12mg with one step every four weeks 8.
V · The 12mg maintenance dose is a community norm
A 12mg weekly maintenance dose is a legitimate community-reported norm with a straightforward rationale, and it sits inside the tested trial range.
The assumption that maintenance has to be a low dose is the most common reason people taper their way out of a good result, and the community rejects the idea outright: “Just don’t assume that the maintenance dose will necessarily be low, it will be whatever it needs to be to keep you on target” [anecdotal] 8. The companion position is that a maintenance dose only has to hold the weight, and it carries no obligation to be small, since the requirement is effectiveness at holding the line rather than economy of milligrams 8.
The norm that follows from that principle is a 12mg weekly maintenance dose, and the most experienced voices in the corpus run exactly that.
“I’ve continued at 12mg since reaching my target weight. I could probably drop the dose down but 12 is working well, I have no issues eating enough, and it’s cheap so why not?”
Community member, Retatrutide protocol corpus, 2026
Another voice describes the same dose in the same tone: “12 is a good maintenance dose for me. I find it easy to run a calorie deficit, surplus, or maintain weight at that dose” [anecdotal] 8. The choice reads as laziness only if the drug is treated as a diet aid, and the community treats it as a chronic therapy instead.
The rationale for staying at 12mg starts with speculation, since one commenter guessed that cardiovascular outcomes, which the registration trials are designed to measure, might track with dose, although no trial has reported such results [anecdotal], and continues with hedonic control, since a low maintenance dose may not suppress the pull of food the way the full dose does 8. On top of that, 12mg is the top of the phase 3 envelope, so the norm sits inside tested territory rather than experimental 1. A maintenance dose of 12mg is therefore a defensible choice with a documented rationale, and the dose a person maintains on says nothing about the quality of their result.
The deeper context is that GLP-1 therapy is chronic, since “They’re designed so that you can continue to use them for the rest of your life” [anecdotal] 8, and the community’s favorite analogy is the blood pressure question: “Do you cycle your blood pressure medication?” [anecdotal] 8. The trial record backs the chronic view, because the majority who stop regain most of the weight, and the community’s summary is blunt: “Most of the folks who stop taking GLP-1s regain most of their weight in a couple years” [anecdotal] 8. Against that background, a 12mg maintenance dose is the unremarkable choice of someone who intends to stay, and the community treats it as such.
The maintenance phase is where the drug keeps doing its work, and the community’s analogy is blood pressure medication, since stopping a GLP-1 returns the defended weight the way stopping an antihypertensive returns the blood pressure 8. Regain after stopping is the expected default, so the dose that holds the line is doing its job for as long as the therapy runs.
VI · The evidence on breaking a plateau
The plateau responds to dose and training, whereas planned breaks and receptor resets rest on no evidence at all.
The honest options when the scale holds for six weeks reduce to escalating within the tested envelope or holding steady while intake and training get a hard look, and the option that fails is the break. The community is flat about it: “There is no science showing any benefit to cyclic GLP-1 therapy” [anecdotal] 8, because a reset cannot reset a receptor that has kept responding 5. Cycling off resets nothing except the GI side effects, and restarting after a long break can backfire, since the corpus records people who came back needing higher doses than the ones they left 8.
Escalation is the evidence-backed move for a stall with room on the ladder, because “GLP-1s support a finite amount of weight loss at a given dose” [anecdotal] 8, and the phase 2 fast-escalation arm shows the value of getting there, with the group that started at 4mg and moved to 8mg at four weeks still losing weight at 44 weeks 7. Total dose reached matters more than the speed of titration, since the only thing that seems to matter for total weight loss is reaching a dose sufficient to the goal, and the timing is secondary 8. The non-responders get the same push: “Try 4, try 6. Not everybody gets a response at low doses” [anecdotal] 8.
Holding is the evidence-backed move at the top of the range, because at 12mg with no side effects and a comfortable maintenance, the plateau is the goal, and the veterans of the community count a held weight as a win, since the maintenance veteran’s line is that holding steady through a vacation counts as a victory [anecdotal] 8. Expectations matter too, because “you might never get that ‘magical weight loss’ effect and it’ll be a grind the whole way down” [anecdotal] 8, and a plateau at the top of the range with good labs is a legitimate end state.
The last lever is what the weight is made of, since “your calorie deficit will determine how much weight you lose but your effort in the gym is the main determinant of what you’re actually losing” [anecdotal] 8, and protein at about 1g per pound of goal weight becomes the floor during reta’s glucagon-driven amino acid catabolism 8. The community’s shortlist of what works is four items: a GLP-1 with amylin if needed, resistance training, protein, and testosterone when it makes sense [anecdotal] 8, and that list rather than a drug holiday is the honest plateau protocol.
The open questions are real, since the retatrutide trials are young and the plateau behavior at four and five years is still being measured, and the phase 3 follow-up will show whether reta’s curve bends the way tirz’s did 2. Until then, the community’s working answer is the one this article has been circling: the plateau is the drug working, and the maintenance dose is where the drug does its work. None of this is medical advice, and every compound mentioned here is research use only.
A break does not restore receptor sensitivity, since the receptor has kept answering, and the trial record shows rebound rather than benefit after discontinuation 5 8. The honest play at a stubborn plateau is dose work and protein, with training carrying the body-composition side of the ledger.
- Jastreboff, A. M., et al., “Retatrutide, a GIP, GLP-1, and glucagon receptor triagonist for the treatment of obesity,” New England Journal of Medicine, 2023. Mean weight loss at 48 weeks was 8.7% at 1mg and 24.2% at 12mg, and the highest-dose group was still losing weight when the trial ended.
- Jastreboff, A. M., et al., “Tirzepatide once weekly for the treatment of obesity,” New England Journal of Medicine, 2022 (SURMOUNT-1). About half of participants had reached their plateau by weeks 24 to 36, close to 90% by week 72, and the extension followed participants to 176 weeks.
- Ravussin, E., et al., metabolic-ward study of tirzepatide, Cell Metabolism, 2025, as summarized in the community corpus. The treated group lost about 38.5 lbs versus 18.5 lbs on placebo, with total daily energy expenditure down about 414 kcal versus 197 and sleeping metabolic rate down about 204 kcal versus 94.
- Hall, K. D., and colleagues, metabolic-ward energy-balance studies, 2021-2022, as cited in the community corpus. The plateau matches a model in which returning appetite closes most of the gap between intake and output, with a smaller contribution from reduced energy expenditure.
- PMC6155907, as referenced in the community corpus: GLP-1 receptor internalization reaches steady state within hours of the first dose, with no progressive long-term desensitization.
- PMC11132924, as referenced in the community corpus: energy-balance modeling shows increased intake dominating reduced energy expenditure at plateaus.
- Phase 2 retatrutide dose-response, as summarized in the community corpus: the 2-4-8mg escalation group lost 21.7% at 48 weeks versus 24.2% for the 2-4-8-12mg group.
- Community corpus: patterns and quotes labeled [anecdotal] are compiled from retatrutide protocol discourse across ret_1 to ret_8 and cluster_research_biohacking (Reddit research-peptide community, 2025-2026). Single-case outcomes are anecdotes, not evidence, and all compounds discussed are research use only.