MK-677 — the oral ghrelin mimetic Merck shelved and the community kept alive

The question the team was working toward was whether you could activate the growth hormone axis orally. Whether a molecule could be designed that didn't behave like a peptide — something the gut would destroy — but behaved like one at the receptor, binding the ghrelin receptor in the hypothalamus and pituitary and producing the same downstream cascade: endogenous GH release, sustained IGF-1 elevation, tissue repair, lean mass preservation. A pill that did what the injections did, without the injections.

MK-677, later called ibutamoren and assigned the research designation MK-677 or L-163,191, was the compound they landed on. It is not a peptide. It is a non-peptide small molecule — a spiropiperidine with a structure entirely unlike any natural hormone — that nonetheless binds the growth hormone secretagogue receptor (GHS-R1a) with high affinity and activates it with high oral bioavailability. It survives digestion. It reaches systemic circulation. And it does what the researchers hoped: it stimulates the pituitary to release growth hormone. Not by delivering GH directly. By triggering the body's own release machinery.

The receptor it targets, GHS-R1a, is the ghrelin receptor. Ghrelin is the hunger hormone — it's what makes your stomach growl and your appetite spike before a meal — but it does far more than that. Ghrelin activating GHS-R1a in the hypothalamus and pituitary is one of the primary signals driving endogenous GH pulses. MK-677 mimics ghrelin at that receptor and holds the signal open longer than the endogenous ligand would. The result is amplified GH release and, over weeks of use, sustained IGF-1 elevation that reflects the cumulative increase in GH output.

This was a genuinely elegant solution to the adherence problem. The pharmacokinetics worked: oral bioavailability in the range of 60-80%, a half-life of four to six hours, once-daily dosing. And the early trials showed the biomarker effects you'd hope for. Elderly subjects showed meaningful IGF-1 increases. Lean mass markers improved modestly. And there was something the researchers didn't necessarily expect: improvements in slow-wave sleep and sleep architecture. The GH axis and sleep are coupled — GH is released primarily during the first slow-wave pulse of the night — and something that amplifies that axis was apparently amplifying the sleep architecture that goes with it.

Merck ran serious trials. The AGHD (Adult Growth Hormone Deficiency) and frailty programs produced data on hip fracture recovery, sarcopenia in the elderly, and markers of metabolic and body composition change. One particularly well-cited study followed elderly subjects over twelve months and showed sustained IGF-1 elevation, modest improvements in lean mass, and some improvements in functional measures — along with the sleep architecture findings that would later attract a separate line of research. The compound was working on the markers it was supposed to work on.

And then the picture got more complicated, in the way that pharmaceutical development tends to get more complicated in Phase II and III.

The fluid retention was real and not trivial. GH elevation, even endogenous GH elevation, increases aldosterone and promotes sodium reabsorption. In elderly people with cardiac or renal considerations, this was a meaningful signal. Ankle edema was reported. The water weight that MK-677 produced looked, superficially, like lean mass gain on certain measurement methods — which created noise in interpreting body composition outcomes. Insulin sensitivity also showed signs of being affected. IGF-1 elevation and the metabolic effects of sustained ghrelin receptor activation both put pressure on glucose handling. In subjects who already had pre-diabetic markers, this wasn't a trivial concern. In an elderly frailty population — the exact population the compound was designed for — metabolic comorbidities are nearly universal.

The clinical endpoints also didn't land the way the biomarker endpoints did. IGF-1 went up, consistently. Lean mass markers improved, modestly. But the functional outcomes — mobility, falls, quality of life, strength — were more equivocal. The effect sizes in hip fracture recovery, one of the target indications, were not compelling enough to build a regulatory case on. The metabolic concerns were not disqualifying but they were real. The combination — modest functional benefit, real side effect signal, an elderly population with narrow safety margins — contributed to a decision that pharmaceutical companies make constantly and never announce as failures: they quietly walked away.

This was not a safety scandal. MK-677 was not withdrawn because of catastrophic adverse events. There is no story of harm here in the dramatic sense. What happened is the more common pharmaceutical story: a compound that worked on mechanisms but not outcomes, in a population that was more complicated than the mechanism suggested, with a side effect profile that made the risk-benefit picture unconvincing for the indications Merck was pursuing. The program was discontinued. The patents eventually aged out of their most commercially valuable period. MK-677 became, in the regulatory sense, a compound without a home.

What happened next is what makes the story unusual. Because MK-677 did not disappear.

The bodybuilding community had been watching growth hormone secretagogues since the late 1990s. Injectable peptides — Sermorelin, GHRP-2, GHRP-6, later Ipamorelin — were circulating in performance and physique culture, used for GH axis support with the same logic that drives most peptide use in that world: the regulatory pathways for approved GH are narrow, the cost is prohibitive, the prescribing requirements are strict. Secretagogues offered a route to GH axis support that was, at various points and in various jurisdictions, more accessible. And when MK-677 appeared — oral, convenient, with a long half-life and a clear mechanism — it was adopted quickly. Not from the pharmacy. From research chemical suppliers. In doses that didn't necessarily match the clinical trials. For purposes that definitely didn't match the frailty and hip fracture programs Merck had been running.

By the 2010s, MK-677 had a substantial online community. Reddit threads, forum archives, YouTube channels, longevity discussion boards — detailed, often technically sophisticated discussions of dosing, cycling, side effects, and lab results. People were testing their own IGF-1 levels before and after. Sharing sleep tracker data. Comparing water retention timelines. The community was doing something the pharmaceutical program never quite did: running a continuous informal observational study at large scale, in young and middle-aged people, at doses both higher and lower than the clinical trials used, with self-reported outcomes that covered the full range from enthusiastic endorsement to cautionary disappointment.

The longevity community found MK-677 through a different door. The interest in GH axis support as a potential aging intervention — not for muscle, but for the slow-wave sleep coupling, the tissue repair downstream, the body composition maintenance that declines steadily from the forties onward — made MK-677's oral convenience genuinely appealing. Injectable secretagogues require subcutaneous administration. MK-677 does not. For someone interested in the mechanism but not interested in daily injections, that matters.

What resulted is a compound with a peculiar status: extensively studied in clinical trials that showed real biomarker effects and insufficient clinical outcomes to justify approval; not FDA-approved for any use in humans; widely used in a community that is neither the frail elderly population nor the elite bodybuilders it is sometimes associated with; and almost entirely outside the medical mainstream, which means that most people using it are doing so without the labs, the metabolic monitoring, or the clinical oversight that would make the risk picture legible.

The Merck program produced something valuable that the company chose not to commercialize. The community found the residue of that decision and ran with it. That gap — between what the trials demonstrated and what the regulatory outcome was — is exactly where the conversation around MK-677 now lives. Not in a prescriber's office, mostly. Not in an approved indication. In forums, in stacks, in individual experiments with IGF-1 panels and sleep trackers and the peculiar democratic science of people who decided to figure it out themselves.

Whether that is a good thing is genuinely complicated. The mechanism is real. The biomarker effects are real. The risks — metabolic, particularly the insulin sensitivity question; cardiovascular in people with existing concerns; and the unknowns that come with long-term use that no clinical trial ever studied — are also real. What Merck walked away from was not a proven therapy. It was a compound that showed enough promise to run serious trials on and not quite enough clinical evidence to take across the regulatory finish line. The community that kept it alive is running on that ambiguity. The compound has not gotten safer or better-evidenced in the years since the trials. It has gotten more accessible. Those are not the same thing.