MK-677 entered Phase 3 this year, in children. In adults, FDA lists it as a safety risk.

Why this note exists

Most compounds in this market are short of human evidence. MK-677 has the opposite problem: there is a substantial randomized record, it is thirty years deep, and almost none of it asks the question the research market is actually interested in. Reading it honestly means reporting a genuine positive alongside a genuine safety signal alongside a live Phase 3 in an unrelated population.

In the index generated on August 15, 2026, 12 of 112 vendors carried an MK-677 listing — 28 listings in total, sixty-second of 83 compounds by vendor coverage.[19] Four of those 28 name ibutamoren anywhere in the product title.[19] Eight of the 28, across two vendors, are not single-compound products at all but multi-compound bundles pairing MK-677 with RAD-140, LGD-4033, ostarine, YK-11, GW-501516 or SR-9009.[19] Where a form is named it is a capsule, a tablet, an oral liquid or a bulk powder, which matters more than it sounds: every human trial below dosed it by mouth, so the route people use and the route that was studied are for once the same one.

A note on what kind of evidence this is: every efficacy and safety figure in this note is human data, carrying its design and participant count. Open-label, single-arm and uncontrolled results are identified as such in the same sentence as the number. One paragraph reports a single case report and calls it that. Doses appear only as their sources published them.

The identity

Three names, one molecule, and it is not a peptide

MK-677, MK-0677, ibutamoren and LUM-201 are the same compound. The clearest statement of that is in a 2022 paper in Hormone Research in Paediatrics, whose first sentence reads: "LUM-201 (ibutamoren, formerly MK-0677) is an orally administered GH secretagogue receptor agonist under development for treatment of pediatric growth hormone deficiency."[7] pepmg's registry lists ibutamoren as an alias of MK-677 for the same reason.[19]

What it is not is a peptide, and this is worth saying on a site that indexes peptides. The 1998 trial report that first tested it against protein catabolism describes it in its own methods as "an orally active nonpeptide mimic of GH-releasing peptide."[6] The 1995 PNAS paper that introduced the molecule characterizes it as belonging to a newly synthesized structural class, "mechanistically indistinguishable from the GH-releasing peptide GHRP-6 and the prototypical nonpeptide GH secretagogue L-692,429 but clearly distinguishable from the natural GH secretagogue, GH-releasing hormone."[8]

That distinction is not pedantry. It changes the route, it changes what a vial of powder is for, and it means the immunogenicity and peptide-impurity concerns FDA raises about the injectable GHRPs are not the concerns that apply here.[1] The concern FDA raises about this molecule is a different one entirely, and it is the subject of the section below.

The new development

What entered Phase 3 in May, and what its Phase 2 showed

The registry entry is specific. NCT06948214 is "A Multicenter, 12-Month, Randomized, Double Blind, Placebo-Controlled Phase 3 Efficacy and Safety Study of Daily Oral LUM-201 in Naïve-to-Treatment, Prepubertal Children With Growth Hormone Deficiency," sponsored by Lumos Pharma, triple-masked, with an estimated enrollment of 150 and an actual start date of May 20, 2026.[9] The stated intervention is LUM-201 at 1.6 mg/kg/day, administered orally once daily.[9] A long-term safety extension, NCT07129759, is registered and not yet recruiting.[20]

The Phase 2 behind it has posted results, and they are more interesting than a press line would be. OraGrowtH210 was a 24-month randomized open-label study with an active control — recombinant human growth hormone by injection — enrolling 104 children.[10] At six months, the co-primary comparison of annualized height velocity landed like this:

Mean annualized height velocity, day 1 to month 6, as posted to the registry for NCT04614337.[10]

Read across, the shape is legible: the oral drug produced real growth, and less of it than the injection it was measured against, in a trial where nobody was blinded. That is a reasonable basis on which to run a placebo-controlled Phase 3 — an oral alternative that works less well than a daily injection can still be worth having for a child who has to take it for years. It is not a result about adults, muscle, sleep or body composition, and the Phase 3 is not testing any of those.[9]

One thing this note will not do is bridge the two dose regimes. The adult trials below used a flat 25 mg per day by mouth; the Phase 3 uses 1.6 mg/kg/day in children.[2][9] A per-kilogram dose and an absolute dose are different kinds of record, and pepmg reports each as its source published it rather than converting one into the other.

Evidence ledger

The adult record, trial by trial

Four randomized adult trials carry most of the weight here. Three of them are placebo-controlled and two are large. They agree with each other about the biomarker and disagree with the marketing about everything downstream of it.

Alzheimer's disease — 563 patients, twelve months, no clinical effect

This is the largest trial the molecule has ever had, and it was run by the company that discovered it. A double-blind multicenter study randomized 563 patients with mild to moderate Alzheimer's disease to MK-677 25 mg or placebo daily for 12 months; 416 completed.[3] Target engagement was not in doubt: serum IGF-1 rose 60.1% at six weeks and 72.9% at twelve months.[3] On outcomes, the paper reports "no significant differences between the treatment groups" on the clinician's impression of change, or in mean change from baseline on ADAS-Cog, ADCS-ADL or CDR-sum of boxes.[3] The authors' conclusion is unhedged: "despite evidence of target engagement … MK-677 25 mg was ineffective at slowing the rate of progression of Alzheimer disease."[3]

One discrepancy worth flagging rather than smoothing over: the publication reports 563 patients randomized, while the registry record for protocol 0677-030 lists actual enrollment of 512.[3][13] The public records do not explain the difference, and this note does not guess at it.

Healthy older adults — two years, fat-free mass up, strength and function unchanged

This is the trial the body-composition claim actually rests on, and it is a good one: a two-year, double-blind, randomized, placebo-controlled, modified-crossover study in 65 healthy adults aged 60 to 81, taking oral MK-677 25 mg or placebo once daily, with fat-free mass and abdominal visceral fat as the primary endpoints at one year.[2]

As reported in the trial publication; body weight rose 2.7 kg on MK-677 against 0.8 kg on placebo (P = 0.003).[2]

Two more findings from the same paper belong in the same breath. Fasting blood glucose rose an average of 0.3 mmol/L, which the authors give as 5 mg/dL (P = 0.015), and insulin sensitivity decreased.[2] The most frequent side effects were an increase in appetite that subsided within a few months, and transient mild lower-extremity edema and muscle pain.[2]

And the authors' own limitation is the load-bearing sentence for anyone reading this as a muscle study: "Study power (duration and participant number) was insufficient to evaluate functional end points in healthy elderly persons."[2] That is an absence of evidence about strength, not a demonstration that there is no effect on it. The honest reading is that after two years in 65 people, the trial could measure the tissue change and could not measure whether it did anything.

Obese men — eight weeks, and a glucose signal

An earlier randomized, double-blind, parallel, placebo-controlled trial gave MK-677 25 mg or placebo daily for eight weeks to 24 obese men aged 18 to 50, twelve per arm.[5] IGF-1 rose about 40% (P < 0.001) and fat-free mass increased significantly by two independent measurement methods, while total and visceral fat did not change significantly.[5] Basal metabolic rate was up at two weeks (P = 0.01) but not at eight (P = 0.1).[5] Fasting glucose and insulin were unchanged, but an oral glucose tolerance test "showed impairment of glucose homeostasis at 2 and 8 weeks."[5] Two trials, twenty-five years apart, pointing the same way on glucose.

Caloric restriction — eight people, and the cleanest positive in the file

The strongest short-term positive result is also the smallest study. A double-blind, randomized, placebo-controlled, two-period crossover trial put eight healthy volunteers aged 24 to 39 on 18 kcal/kg/day for two 14-day periods, adding oral MK-677 25 mg or placebo for the last seven days of each.[6] Mean daily nitrogen balance in the treatment week was +0.31 g/day on MK-677 against −1.48 g/day on placebo (P < 0.01) — a reversal of diet-induced protein catabolism, not merely an attenuation of it.[6]

It is a real, well-designed, mechanistically coherent finding. It is also eight people, for seven days, under severe caloric restriction, measuring a nitrogen balance rather than anything a person would notice, and it has not been followed by a trial that turned it into a functional outcome.[6]

The safety signal

The trial that stopped, and the list it put the compound on

In 2011 a multicenter, randomized, double-blind phase IIb study reported on 123 elderly patients recovering from hip fracture, assigned to 25 mg/day of MK-0677 (n = 62) or placebo (n = 61).[4] IGF-1 behaved as it always does, rising 51.4 ng/mL against placebo (95% CI 34.42 to 68.44; P < 0.001).[4] The primary functional measure did not: mean stair climbing power at 24 weeks rose 12.5 W against placebo with a confidence interval spanning zero (95% CI −10.95 to 35.88; P = 0.292).[4] Gait speed did improve (P = 0.011), and the authors record that there was "no improvement in MK-0677 treated patients in several other functional performance measures."[4]

Then the sentence that outlived the trial: "Trial was terminated early due to a safety signal of congestive heart failure in a limited number of patients."[4] The authors' conclusion is equally plain: "MK-0677 has an unfavorable safety profile in this patient population."[4] The published abstract does not give per-arm event counts, and this note does not supply numbers its source did not print.

That trial is the reason ibutamoren appears on an FDA list at all. In the agency's interim policy on compounding, ibutamoren mesylate sits in category 2 — bulk drug substances that may present significant safety risks — under both 503A and 503B, dated September 29, 2023 and December 29, 2022 respectively.[1] FDA's stated basis, on a page whose content is current as of April 22, 2026, is worth quoting in full because it is two sentences long:

Two things follow from that, and they pull in different directions. A category 2 listing is a statement about compounding eligibility, not a finding that the drug caused heart failure in anyone — FDA's own word is "potential," and the trial's word is "signal."[1][4] But it is also the agency, in writing, naming a specific cardiac concern about this specific molecule, which is a materially different position from having no opinion. And the population it was seen in — elderly patients recovering from hip fracture — is not the population buying it.

The registry adds a loose end this note cannot close. It carries a separate terminated post-hip-fracture study of 83 participants under protocol 0677-032, completed in August 2007, whose public title reads only "Proprietary Information - Exploratory (Non-Confirmatory) Trial" and which states no reason for termination.[21] Whether that is the trial published in 2011 or a second one, the public records do not say, and pepmg is not going to assume.

The most recent human data goes the wrong way, in seven people

The newest completed adult study on the registry is not an industry trial. It is an investigator-initiated pilot at Massachusetts General Hospital, "The Impact of Ibutamoren on Nonalcoholic Fatty Liver Disease," completed on December 23, 2024.[11] The design is the important part: phase 2, single-group, open-label, with historical controls — no randomization and no concurrent comparison.[11] Twelve participants enrolled, seven completed, and the posted outcomes are analysed on those seven.[11]

The hypothesis was that LUM-201 would decrease intrahepatic lipid.[11] The posted results at six months move the other way on all three measures: mean intrahepatic lipid content rose 3.9 percentage points (SD 6.7), the LiverMultiScan corrected T1 score rose 39 ms (SD 44), and ALT rose 8 U/L (SD 18).[11] No serious adverse events were posted.[11]

Seven uncontrolled participants cannot establish harm any more than they could have established benefit, and standard deviations that large across that few people are consistent with almost anything. But it is the freshest human read on the compound, it was run by academics with no product to sell, and it did not find what it went looking for. An index that would have reported a 3.9-point fall as encouraging has to report a 3.9-point rise.

The question the market cares about is registered, and has not started

There is one trial in the registry aimed squarely at the reason people buy this compound in 2026 — preserving muscle and function while losing weight on a GLP-1. NCT07754045 is a phase 2, randomized, double-blind, placebo-controlled, multi-centre study "evaluating the efficacy and safety of oral LUM-201 as an adjunct to oral semaglutide for improving physical function in older adults with obesity," in 202 adults aged 60 to 85 with a BMI between 30 and 45 and mild functional impairment.[12]

Its status as of a registry query on August 27, 2026 is not yet recruiting, with an estimated start date of February 15, 2027.[12] No participant has been dosed. A registered trial is a hypothesis with a budget attached, not a result, and until it reports there is no human evidence that MK-677 preserves function during GLP-1 weight loss — in either direction.

For scale on the whole file: a ClinicalTrials.gov search for interventional studies of ibutamoren on August 27, 2026 returned 8 records, and a search on the LUM-201 development name returned 8, of which 2 appear in both — 14 distinct registered studies across thirty years.[14][23]

Status elsewhere: unapproved, and prohibited in sport

There is no approved product. A Drugs@FDA query on August 27, 2026 returns no approved drug with ibutamoren as an active ingredient, which is consistent with the compound being in Phase 3 rather than on the market.[15]

In sport it is named explicitly. The 2026 Prohibited List places it under S2.2.4, growth hormone releasing factors, in the clause covering "growth hormone secretagogues (GHS) and their mimetics (e.g. anamorelin, capromorelin, ibutamoren (MK-677), ipamorelin, lenomorelin (ghrelin), macimorelin and tabimorelin)."[16][17] That is prohibited at all times, in and out of competition.[16]

It is also detectable long after the fact. A 2026 forensic toxicology paper reports that after a single 10 mg oral dose given to a male volunteer, ibutamoren was still detectable in the first centimetre of hair collected four weeks later, at 1.3 pg/mg against a limit of detection of 0.1 pg/mg.[18] For anyone subject to testing, "I only tried it once" is not a defence a hair sample supports.

One case report, labelled as one

A March 2026 case report in Cureus describes a 54-year-old man who presented with atraumatic splenic rupture after recent use of MK-677 and RAD-140 for bodybuilding, requiring embolization and then emergency splenectomy; histopathology showed splenic infarction "with features raising the possibility of an underlying vascular malformation."[22]

The authors are careful, and this note will be no less careful than they were. They describe their own reasoning as exploring "the potential and speculative role of performance-enhancing compounds," note that "RAD-140 and MK-677 have not been previously linked to splenic pathology," and raise a pre-existing vascular malformation as a candidate explanation.[22] One patient, two compounds, a plausible alternative cause, and no control group is a report of an event, not evidence of a hazard. It is here because it exists and because leaving it out of a safety section would be a choice too.

Three things this note is not saying

First, it is not saying MK-677 does nothing. It reliably raises growth hormone and IGF-1 in humans, it reliably increases fat-free mass across three independent randomized trials, and it reversed nitrogen loss under caloric restriction.[2][5][6] Those are consistent, replicated human findings, and they are the reason the molecule is still in development.

Second, it is not saying the compound causes heart failure. FDA wrote "potential," the trialists wrote "signal," the abstract gives no per-arm counts, and the observation comes from frail elderly patients recovering from a fracture.[1][4] What it is saying is that an early-terminated trial and a standing FDA safety listing exist, and that a compound with those attached is in a different category from one with no findings at all.

Third, it is not saying the Phase 3 will fail. The Phase 2 produced a real growth response at the dose being carried forward, and an oral option that underperforms a daily injection can still be the right trade for a child.[10]

What it is saying is narrower. Thirty years of human trials have established what MK-677 does to a blood test and to a DXA scan, and have not established that it changes strength, function, cognition, liver fat or recovery in adults — while two of them found glucose moving the wrong way and one of them stopped early.[2][3][4][5][11] The trial that would test the thing people actually want is scheduled to begin in February 2027.[12]

Questions people are asking