Sermorelin

Chemical identity

Sermorelin is registered twice, once as the free peptide and once as the acetate, under different CAS numbers. PubChem resolves the name to one record: CID 16132413, CAS 86168-78-7, formula C149H246N44O42S, molecular weight 3357.9 g/mol, InChIKey WGWPRVFKDLAUQJ-MITYVQBRSA-N, UNII 89243S03TE. The FDA Global Substance Registration System holds both, the free peptide under that same UNII and sermorelin acetate under UNII 00IBG87IQW with CAS 114466-38-5. Either way the sequence is the same twenty-nine residues, YADAIFTNSYRKVLGQLSARKLLQDIMSR, with a C-terminal amide. That is human growth hormone-releasing hormone truncated after residue 29, with no substitution anywhere in the chain. Prakash and Goa's 1999 review describes it as the shortest synthetic peptide with the full biological activity of GHRH (PMID 18031173).

One conflict is recorded here and left unresolved. GSRS displays a molecular formula of C149H247N44O43S alongside a sequence-calculated mass of 3357.91. Those two figures do not reconcile: the formula shown carries an extra hydrogen and an extra oxygen and would compute to roughly 3375. PubChem's C149H246N44O42S computes to 3357.9, which matches the mass GSRS itself displays. The same pattern appears on the GSRS record for tesamorelin, where a calculated formula omits the amidation described in the record's own structural notes. Both formulas are published here. The identity block carries PubChem's, because it agrees with the mass that both registries report.

The approval that was withdrawn

Two US approvals existed. NDA 19-863, a 0.05 mg base ampoule, was approved on 28 December 1990 and indicated for evaluating the ability of the pituitary somatotroph to secrete growth hormone. NDA 20-443, at 0.5 and 1.0 mg base per vial, was approved on 26 September 1997 and indicated for idiopathic growth hormone deficiency in children with growth failure. Both applications were held by the same manufacturer. Drugs@FDA records the 1997 approval as a Type 3 new dosage form under standard review with orphan status.

The manufacturer requested the withdrawal. FDA did not act on safety or effectiveness grounds. A letter dated 11 July 2008 notified FDA that the diagnostic ampoule was being discontinued, and letters dated 2 and 12 December 2008 requested withdrawal of both applications. FDA announced the withdrawal of approval of NDA 19-863 and NDA 20-443 effective 18 June 2009. A citizen petition filed in October 2012 prompted a determination, published on 4 March 2013, that neither product was withdrawn from sale for reasons of safety or effectiveness, which keeps both in the Discontinued Drug Product List section of the Orange Book and leaves the applications open to generic reference.

What is no longer retrievable matters for anyone trying to check a figure. The Drugs@FDA entry for NDA 20-443 states that the label is not available on that site, so the prescribing information for the treatment product cannot be read there. An FDA-commissioned review completed in December 2020 searched the national medicine registers of thirteen countries and regions and found no approved sermorelin product in the United States or in any of the others, and recorded that no United States Pharmacopeia monograph existed for sermorelin acetate.

What the trials in children measured

The largest treatment study is the multicentre, open-label trial published in 1996 by Thorner and the Geref International Study Group. One hundred and ten previously untreated prepubertal growth-hormone-deficient children received 30 micrograms per kilogram per day subcutaneously at bedtime for up to a year, and 86 were eligible for the efficacy analysis. Mean height velocity moved from 4.1 plus or minus 0.9 cm per year at baseline to 8.0 plus or minus 1.5 at six months and 7.2 plus or minus 1.3 at twelve. Seventy-four per cent were classed as good responders at six months. There was no placebo arm and no randomisation.

Head-to-head comparisons with growth hormone itself came out differently. Neyzi and colleagues in 1993 randomised 43 prepubertal children with growth hormone deficiency of hypothalamic origin to subcutaneous GHRH(1-29)NH2 at either of two doses or to growth hormone for six months; height velocity was comparable in the high-dose and growth hormone groups, but height standard deviation score for bone age rose only in the growth hormone group (PMID 8329826). Ogilvy-Stuart and colleagues in 1997 treated nine children with radiation-induced deficiency for a year, recording height velocity of 3.3 to 6.0 cm per year, then 7.5 cm per year during the following year on growth hormone.

Two findings from this period cut against the idea that the pituitary response holds indefinitely. Kirk and colleagues treated eighteen short prepubertal children who were not growth-hormone-insufficient for twelve months; height velocity rose from 4.8 to 7.2 cm per year, but overnight growth hormone levels and growth hormone responses to GHRH testing both fell across the treatment year. Tauber and colleagues infused six children with partial deficiency continuously for six months and recorded a biphasic result: 24-hour integrated growth hormone rose from 1.6 to 3.5 micrograms per litre per minute, then fell to 1.9 by six months, with one child showing complete suppression of secretion. Lanes and Carrillo detected antibodies to GHRH in 4 of 11 responders at six months and 6 of 11 at twelve.

How quickly it disappears

Soule, King and Millar measured this directly in 1994. Ten normal men received GHRH(1-29)NH2 by constant intravenous infusion at 25 nanograms per kilogram per minute, with sampling during the ninety-minute infusion and for twenty minutes afterwards. The disappearance half-time was 4.3 plus or minus 1.4 minutes and the metabolic clearance rate 39.7 plus or minus 3.9 mL per kilogram per minute. The same protocol run with the D-Ala2-substituted analogue gave 6.7 plus or minus 0.5 minutes and a clearance rate of 21 plus or minus 1.2, which is the experiment that established what the position-2 substitution buys.

Wilton and colleagues characterised the dose-response and the nasal route in thirty healthy men aged 19 to 43. Intravenous injection at 0.25 micrograms per kilogram, the lowest dose tested, produced significant growth hormone release, with maximal release at 1 to 2 micrograms per kilogram. The peptide was eliminated rapidly, but growth hormone concentrations stayed elevated for about three hours. Absorption across the nasal mucosa was low: bioavailability was 3 to 5 per cent, and roughly 50 micrograms per kilogram intranasally matched 1 microgram per kilogram intravenously.

Where the molecule is cut has been mapped, though in the rat. Boulanger and colleagues incubated rat GRF(1-29)NH2 in rat serum and liver homogenate and reported apparent half-lives of 18 plus or minus 4 and 13 plus or minus 3 minutes; the Ala2-Asp3 bond, the dipeptidyl peptidase IV site, was hydrolysed in both preparations. Memdouh and colleagues later characterised sermorelin(3-29)-NH2, the product of that same cleavage, as an in vitro metabolite, synthesised it as a reference material and spiked it into fortified urine to build a detection method (PMID 34665524). It has not been reported in a real sample. Neither the serum nor the liver figure is a human plasma half-life.

Adults, and the analogue mistaken for it

Corpas and colleagues at the National Institute on Aging studied ten men aged 68.0 plus or minus 6.2 alongside nine young men in 1992. The older men took low and high doses subcutaneously twice daily for fourteen days each, separated by a fourteen-day washout. Increases in mean 24-hour growth hormone, area under peaks, peak amplitude and IGF-I reached significance against the old baseline only at the higher dose. After that dose, none of those measures differed between the age groups. Fasting glucose, urinary C-peptide, blood pressure and chemistry profiles were unaffected.

Vittone and colleagues ran the nightly-dose version in 1997: eleven healthy non-obese men aged 64 to 76 with low baseline IGF-I, self-injecting at home for six weeks. Mean nocturnal growth hormone release, area under the growth hormone peak and peak amplitude all rose. IGF-I, IGF binding protein-3 and growth hormone binding protein did not. Weight, body mass index, waist-to-hip ratio and dual-energy x-ray absorptiometry measures of muscle and fat did not change, nor did muscle histology or glucose and insulin responses. Two of six strength measures and one endurance test improved. The authors concluded that single nightly doses were less effective than multiple daily doses.

Most of what circulates about adults traces to a study of a different peptide. Khorram, Laughlin and Yen enrolled ten women and nine men aged 55 to 71 in a five-month single-blind placebo-controlled trial and gave sixteen weeks of [Nle27]GHRH(1-29)NH2, a norleucine-substituted analogue, at 10 micrograms per kilogram nightly (PMID 9141536). Skin thickness increased in both sexes, lean body mass in men only, and general well-being and libido in men only. Sleep quality was unaffected in both. The peptide is not sermorelin; the substitution is named in the paper's own title.

Vitiello and colleagues published a cognition result in 2006 from eighty-nine healthy older adults, mean age 68.0 plus or minus 0.7, assigned to six months of daily treatment or placebo. MEDLINE indexes the report as a controlled clinical trial and not as a randomised controlled trial, and the paper itself does not describe randomisation. Performance improved on WAIS-R performance IQ, WAIS-R picture arrangement, finding A's, verbal sets and a single-dual task. The 2006 abstract names only GHRH. A companion review from the same group identifies the drug used in both of its trials as GHRH(1-29)NH2, sermorelin acetate, given as a single evening subcutaneous injection at 14 micrograms per kilogram. The molecule administered is therefore documented, which is unusual in the adult literature.

Sleep: registered, then not reported

A trial was registered for exactly this question and its sleep result was never published. NCT00000380, run at the University of Washington with the National Institute of Mental Health, is titled Growth Hormone Releasing Hormone (GHRH) Treatment for Age-Related Sleep Disturbances. It started in June 1996, is listed as completed in July 2007, and has no results posted. The registry links two result publications: the 2006 cognition paper, and a 2004 analysis of baseline subjective and objective sleep quality in 150 healthy older adults (PMID 15172206). Neither reports a treatment effect on sleep; the 2004 paper is a cross-sectional baseline analysis, not an outcome comparison against placebo. A PubMed search returns no polysomnography outcome from the trial.

Interim numbers from that trial went the other way. Reporting preliminary data in 2001 while the study was still running, the investigators recorded that the Pittsburgh Sleep Quality Index total score rose from 4.1 plus or minus 2.8 to 5.41 plus or minus 2.8 in the treated group, thirty-seven subjects, at P less than 0.05, a small but significant move towards poorer subjective sleep, while the placebo group of thirty-eight was unchanged at 4.51 to 4.61. No component score showed a clear effect. The same review recorded that a single large growth hormone burst followed each evening injection, that nighttime pulsatile secretion was not restored, and that late-night secretion was reduced against baseline profiles.

Human work on GHRH and slow-wave sleep exists, but it does not close the gap. Steiger and colleagues gave four hourly intravenous boluses of 50 micrograms to seven male controls and recorded 20.2 plus or minus 6.6 per cent of the night in slow-wave sleep against 14.0 plus or minus 5.6 on placebo (PMID 1361964); the report does not identify which GHRH fragment was used, and PubMed does not index it to sermorelin. Jessup and colleagues ran the converse experiment in 2004, infusing a GHRH antagonist in healthy men for twelve hours: the growth hormone response to a GHRH bolus was suppressed by 93 plus or minus 1.8 per cent and the percentage of slow-wave sleep was unchanged (PMID 15538933). Merriam and colleagues, reviewing their own programme in 2003, wrote that chronic treatment with a short-acting GHRH did not improve sleep (PMID 14610297).

The compounding record and the doping record

Sermorelin acetate sits in category 1 of FDA's list of bulk drug substances nominated for compounding by outsourcing facilities under section 503B, in the version updated 21 March 2025, carrying the two asterisks that designate a component of an FDA-approved drug. The footnote attached to that designation states that FDA does not intend to act against an outsourcing facility for compounding what is essentially a copy of an approved drug that has been discontinued and is no longer marketed. It appears nowhere in the parallel list for compounding under section 503A, updated 14 May 2026, in category 1, 2 or 3.

How the substance reached that list is documented. An FDA-commissioned report completed in December 2020 records that sermorelin acetate was nominated by a compounding pharmacy and by an outsourcing facility trade body, for growth hormone deficiency and adult-onset growth hormone deficiency, and that the stated reason was the absence of an approved product. The report's literature review searched to January 2020, screened 466 titles, read 65 full texts and included two: both 1996 Italian diagnostic studies, 435 patients between them, one testing arginine with GHRH and one comparing arginine-with-GHRH against pyridostigmine-with-GHRH and IGF-I. Neither examined treatment.

Anti-doping analysis supplies the other regulatory datum. Memdouh and colleagues at King's College London state that GHRH and its synthetic analogues are prohibited by the World Anti-Doping Agency, and name sermorelin as one of four analogues they characterised in fortified urine (PMID 34665524). Their paper records that despite admissions and intelligence indicating use, the compounds do not appear to have been found in samples by laboratories accredited to that agency, and attributes this to low urinary concentrations and limited knowledge of their metabolism. They synthesised nineteen in-vitro metabolites as reference materials and reached limits of detection at or below the 1 ng/mL performance requirement.

What is not known

No randomised placebo-controlled trial has tested sermorelin in adults for a body-composition, strength, sleep or metabolic endpoint and published the result. The adult record is four studies: two of fourteen days and six weeks with hormone endpoints, one cognition trial at six months, and one interim report. The cognition trial is indexed in MEDLINE as a controlled clinical trial and not as a randomised one, so even that result does not fill the gap. The registered trial designed around sleep, NCT00000380, ran from 1996 to 2007 and posted no results; its polysomnography outcome data have never appeared in PubMed, and the two publications the registry links to it report cognition and a cross-sectional baseline sleep analysis. A ClinicalTrials.gov search on sermorelin returns 42 studies, of which eleven match on the intervention field, and not one names sermorelin as the intervention: they are labelled generically as GHRH, or use tesamorelin, GHRH(1-44) or a GHRH-plus-arginine diagnostic test. Only three papers in PubMed carry sermorelin in the title, one of them Prakash and Goa's 1999 review (PMID 18031173) and one an in-silico drug-screening exercise in glioma transcriptomes with no experimental administration. Only one study followed children beyond twelve months: sixteen children treated for 12 to 24 months, with growth velocity reported to 24 months (PMID 7735367). Nothing longer than 24 months in children, or 6 months in adults, was located. The 1999 review states that data in a few children suggest the effect is maintained for 36 months of continued treatment, but no primary report of that follow-up was found. Immunogenicity is characterised by one study in sixteen children. No modern repeat-dose toxicology, carcinogenicity, reproductive or developmental data for the compound could be located, and the prescribing information that would have summarised the registration-era safety file is not available through Drugs@FDA.

Questions