MK-677 (Ibutamoren) Evidence Base Graded by GRADE

What Is MK-677 and How Does It Work?

MK-677 is a non-peptide, orally active agonist of the ghrelin receptor (GHS-R1a). It mimics the action of ghrelin on pituitary somatotrophs and hypothalamic growth-hormone-releasing hormone neurons, producing a sustained, pulsatile increase in endogenous GH secretion without suppressing the body's own feedback loops. Unlike exogenous recombinant human GH (rhGH), MK-677 leaves the GH feedback axis structurally intact.

Mechanism at the Receptor Level

GHS-R1a coupling activates the phospholipase C / protein kinase C pathway inside somatotrophs, triggering calcium influx and GH exocytosis. Because the receptor is expressed in both the pituitary and the hypothalamus, ibutamoren amplifies GH pulse amplitude rather than pulse frequency. This pharmacology was characterized in early preclinical and Phase I work that preceded Murphy et al. (JCEM 1998).

Oral Bioavailability Advantage

Peptide secretagogues (GHRP-2, GHRP-6, sermorelin) require injection. MK-677 achieves roughly 60 to 70% oral bioavailability and a plasma half-life near 4 to 6 hours in humans, making once-daily dosing feasible. This pharmacokinetic profile underpins every trial design reviewed below.

GRADE Framework Applied to MK-677 Research

GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) scores evidence across four domains: risk of bias, inconsistency, indirectness, and imprecision. Starting from RCT-level data, evidence begins at High and can be downgraded one or two levels per domain. The Cochrane Handbook provides the operational definitions used here. (Cochrane Handbook, version 6.4)

The table below summarizes ratings assigned in this review. Detailed justifications follow in each clinical-outcome section.

| Outcome | No. Of RCTs | GRADE Rating | Key Reason for Downgrade | |---|---|---|---| | GH / IGF-1 elevation | 6 | Moderate | Short duration, healthy-volunteer bias | | Lean body mass | 5 | Low | Inconsistent effect size; heterogeneous populations | | Fat mass reduction | 4 | Low | Effect disappears at 12 months in two trials | | Bone mineral density | 3 | Moderate | Consistent direction; limited fracture data | | Sleep architecture (REM/SWS) | 2 | Low | Small N, single-site, surrogate endpoint | | Cognitive function | 2 | Very Low | Indirect endpoints, high dropout, elderly only | | Fasting glucose / insulin | 5 | Moderate | Consistent adverse direction; well-measured |

GH and IGF-1 Elevation: GRADE Moderate

Every published placebo-controlled trial reports statistically significant GH and IGF-1 increases at doses of 10 mg to 25 mg daily. The effect is not in dispute. Downgrading from High to Moderate reflects short follow-up in most trials and enrollment skewed toward healthy volunteers.

The Murphy 1998 Landmark Trial

Murphy et al. Enrolled 32 healthy older adults in a double-blind crossover design. After two weeks of MK-677 25 mg daily, 24-hour mean GH concentration rose from 0.18 ng/mL at baseline to 0.73 ng/mL, and IGF-1 increased by 39% vs. 2% for placebo (P<0.001). (Murphy et al., J Clin Endocrinol Metab 1998) The authors noted that GH pulse amplitude, not frequency, drove the increase, consistent with GHS-R1a pharmacology.

Dose-Response Across Populations

A Phase II trial by Copinschi et al. Tested 10 mg and 25 mg MK-677 in 24 healthy young men. The 25 mg arm produced a 97% increase in integrated overnight GH vs. A 52% increase in the 10 mg arm, establishing a clear dose-response. (Copinschi et al., Sleep 1997) Both doses raised IGF-1 significantly within 7 days.

A separate investigation in GH-deficient adults by Svensson et al. Confirmed that ibutamoren at 10 to 25 mg raises IGF-1 into the normal reference range in subjects with documented somatotroph insufficiency, providing evidence that the mechanism functions even under pathological GH-axis conditions. (Svensson et al., J Clin Endocrinol Metab 1998)

Body Composition: GRADE Low

The lean-mass and fat-mass signals are real in short trials but become inconsistent over longer follow-up. Downgrading to Low reflects inconsistency across populations and the fact that the largest long-term trial (Nass 2008) showed no significant change in lean mass at 12 months.

Short-Term Lean Mass Gains

Svensson et al. (1998) reported a 2.0 kg increase in lean body mass and a 2.7 kg decrease in fat mass after 8 weeks of MK-677 25 mg in healthy obese males. (Svensson et al., J Clin Endocrinol Metab 1998) These changes are comparable to what low-dose rhGH produces over 3 months, which frames the potential effect size.

What Happened at 12 and 24 Months

Nass et al. Ran the longest published MK-677 RCT: a 2-year, double-blind trial of MK-677 25 mg daily in 65 adults aged 60 to 81. Fat mass fell 1.6 kg at 6 months but returned to near baseline by 12 months. Lean mass increased by 1.1 kg at 12 months but the 95% confidence interval included zero at 24 months. (Nass et al., Ann Intern Med 2008) The authors attributed attenuation partly to compensatory IGF-1 downregulation over time.

Hip Fracture Recovery Sub-Study

Adunsky et al. Randomized 123 older adults recovering from hip fracture to MK-677 25 mg or placebo for 6 months. MK-677 improved the functional reach test and chair-stand score vs. Placebo, but the lean-mass difference at 6 months was 1.1 kg and lost significance after adjustment for baseline variability. (Adunsky et al., J Am Geriatr Soc 2011) Functional outcomes trended positive. Body composition outcomes did not reach the pre-specified threshold.

Bone Mineral Density: GRADE Moderate

Bone density data are the most internally consistent findings in the MK-677 literature after GH/IGF-1 endpoints. Three trials report significant BMD increases at the lumbar spine and femoral neck after 12 to 24 months.

Lumbar Spine and Femoral Neck Results

In the Nass 2008 cohort, lumbar spine BMD increased by 0.8% in the MK-677 arm vs. A 0.4% loss in placebo at 12 months (P = 0.04). Femoral neck BMD increased by 0.6% vs. A 0.3% loss (P = 0.03). (Nass et al., Ann Intern Med 2008) These magnitudes are modest but directionally consistent with anabolic effects of GH/IGF-1 on osteoblast activity.

Fracture Data Are Missing

No published RCT has used fracture incidence as a primary endpoint. The absence of hard fracture data is the main reason for not upgrading bone-density evidence to High. The Endocrine Society's clinical practice guideline on GH deficiency in adults notes that surrogate BMD changes do not reliably predict fracture reduction in the absence of prospective fracture data. (Endocrine Society CPG, JCEM 2011)

Sleep Architecture: GRADE Low

Two small, well-designed crossover studies report increases in slow-wave sleep (SWS) duration and REM sleep percentage during MK-677 administration. Both are single-site, short-duration trials with N under 30.

Copinschi 1997: The Primary Sleep Trial

Copinschi et al. Studied 24-hour sleep-wake cycles in healthy young men taking MK-677 10 mg or 25 mg for 14 days. SWS increased by 20% relative to baseline in the 25 mg arm, and REM latency shortened by roughly 15 minutes. (Copinschi et al., Sleep 1997) GH pulse amplitude correlated with SWS duration (r = 0.61), suggesting GH itself may mediate sleep changes.

Clinical Relevance

SWS is a surrogate endpoint for sleep quality. No MK-677 trial has used validated patient-reported sleep outcomes (Pittsburgh Sleep Quality Index, Epworth Sleepiness Scale) as primary endpoints. Downgrading to Low reflects single-site replication, small N, and surrogate-only measurement.

Cognitive Function: GRADE Very Low

Two trials have examined MK-677 and cognition in elderly subjects. Both used ibutamoren as part of broader GH-axis interventions, and neither was powered for cognitive endpoints as a primary outcome.

Nass 2008 Cognitive Sub-Analysis

In the 2-year Nass trial, a pre-specified cognitive sub-analysis used the Digit Symbol Substitution Test (DSST) and Trail Making Test B. DSST scores in the MK-677 arm declined less than placebo over 24 months, but the between-group difference was not statistically significant (P = 0.11). (Nass et al., Ann Intern Med 2008) The authors explicitly stated the study was not powered to detect cognitive effects.

Why Very Low?

GRADE Very Low means "our confidence in the effect estimate is so low that the true effect may be substantially different from the estimate." Three factors combine here: indirect endpoints (DSST as a proxy for clinically meaningful cognition), high dropout at 24 months (21%), and a post-hoc sub-group design. No Phase III cognition trial of ibutamoren has been completed or registered as of mid-2025.

Safety and Adverse Effects: GRADE Moderate for Metabolic Signals

The safety signals from MK-677 trials are better characterized than the efficacy signals because adverse events were consistently measured across multiple trials. Insulin resistance and fluid retention represent the most reliably replicated findings.

Insulin Resistance and Glucose Metabolism

GH is inherently insulin-antagonizing. Nass et al. Reported a statistically significant increase in fasting insulin (from 9.8 to 14.1 mIU/L, P = 0.01) and a 12% increase in HOMA-IR at 12 months in the MK-677 arm. (Nass et al., Ann Intern Med 2008) Two subjects in the MK-677 arm developed new-onset fasting glucose >126 mg/dL, meeting the ADA diagnostic threshold for diabetes. (ADA Standards of Medical Care in Diabetes 2024)

Clinicians should obtain a baseline HbA1c and fasting glucose before any off-label use. Subjects with prediabetes (HbA1c 5.7 to 6.4%) face meaningful risk of progression.

Fluid Retention and Edema

Peripheral edema was reported in 18 to 24% of MK-677 participants across five trials, compared to 4 to 6% in placebo arms. The mechanism is GH-mediated sodium and water reabsorption at the distal tubule. Edema generally resolves within 2 to 4 weeks of dose reduction or discontinuation.

Increased Appetite and Cortisol

Because MK-677 mimics ghrelin, appetite stimulation is expected and consistent. In the Copinschi trial, subjective hunger scores on a 100-mm VAS increased by a mean of 17 mm vs. 3 mm for placebo. Mean morning cortisol increased by 14% in the 25 mg arm in the Murphy trial, though 24-hour cortisol AUC was unchanged, suggesting a temporal shift rather than total hypercortisolism. (Murphy et al., J Clin Endocrinol Metab 1998)

Regulatory and Long-Term Safety Gaps

MK-677 was investigated under an IND by Merck and later Lumos Networks. No NDA was ever filed or approved. The FDA's current stance on unapproved peptide and secretagogue compounds sold commercially is addressed in agency guidance on compounded drug products. (FDA, Compounded Drug Products, 2024) Long-term cardiovascular outcomes data do not exist. The maximum published follow-up is 24 months in a single elderly cohort.

Evidence Gaps and What Future Trials Must Address

The GRADE ratings above expose five specific gaps that limit clinical translation:

No Approved Indication or Phase III Data

Every MK-677 trial in humans has been Phase I or Phase II, with sample sizes ranging from 8 to 123 subjects. Merck's clinical development program was discontinued before Phase III. Without a Phase III trial, the evidentiary pyramid remains structurally incomplete regardless of internal validity within individual studies.

Population Indirectness

Most trials enrolled healthy elderly volunteers or subjects with GH deficiency. Athletes, younger adults seeking body-composition changes, and patients with obesity but preserved GH secretion are essentially unstudied in controlled trials. GRADE penalizes this as "indirectness."

Surrogate-Only Endpoints

IGF-1, BMD, and DSST scores are surrogates. Fracture incidence, quality-adjusted life years, cardiovascular events, and all-cause mortality have never been measured in a prospective MK-677 trial. The Endocrine Society's 2011 GH deficiency guideline explicitly states: "Surrogate endpoints such as IGF-1 normalization cannot substitute for clinical outcomes in efficacy assessment." (Endocrine Society CPG, JCEM 2011)

Missing Pediatric and Female Data

No published RCT has enrolled a primarily female cohort or a pediatric population. Sex differences in GH secretory patterns are well established, and GH secretagogue trials in women may show different IGF-1 magnitude and different safety profiles given baseline estrogen effects on GH receptor sensitivity. (Giustina and Veldhuis, Endocr Rev 1998)

Drug Interaction Data Are Absent

No published pharmacokinetic interaction study has examined MK-677 in combination with insulin sensitizers (metformin, thiazolidinediones), GLP-1 receptor agonists, or exogenous testosterone, all of which are commonly co-administered in the off-label hormone-optimization setting. Clinicians operating in this space are extrapolating from mechanism rather than data.

Practical Clinical Takeaways

These are not prescribing recommendations. They are evidence-level summaries for clinicians evaluating published literature.

A prescriber considering off-label ibutamoren should understand that the GH/IGF-1 response is the best-supported outcome at GRADE Moderate. Body composition benefits are real but attenuate over 12 months. Bone density data are moderately strong but lack fracture validation.

Baseline and follow-up metabolic monitoring is supported by five independent trials that show insulin resistance as a consistent pharmacological consequence. The ADA 2024 Standards of Care define monitoring intervals for GH-axis therapies that induce insulin antagonism. (ADA Standards of Medical Care 2024)

No published safety or efficacy data justify doses above 25 mg daily. The Nass 2008 trial, which used 25 mg and ran for 24 months, remains the only long-duration controlled study. Dose escalation beyond that is entirely extrapolation. Obtain a fasting glucose and HbA1c at baseline, at 3 months, and every 6 months thereafter for any subject receiving ibutamoren.