What it is
GHRP-2 is a synthetic peptide of six amino acids — D-Ala-D-2-Nal-Ala-Trp-D-Phe-Lys-NH2 — whose generic name is pralmorelin. In older papers and patent filings it also appears as GPA-748, KP-102 D and KP-102 LN. PubChem records it under CAS 158861-67-7, formula C45H55N9O6, molecular weight 818.0.
It belongs to a family designed on purpose. In the 1980s and 1990s a group at Tulane University led by Cyril Bowers, working with Polygen in Germany, synthesised a series of small, highly active peptides intended to imitate ghrelin, the 28-amino-acid hormone that tells the pituitary to release growth hormone. GHRP-6 and hexarelin came from the same effort, and GHRP-6 is the same length as this one — six residues, sharing the Trp-D-Phe-Lys tail and differing at the front end. Kaken Pharmaceutical acquired worldwide rights to this one and sublicensed North America to Wyeth.
It is sold today for the reasons anything that raises growth hormone is sold: muscle, recovery, fat loss, sleep. Its actual regulatory history points somewhere else entirely.
It is approved — as a test
The compound's one approval anywhere in the world is in Japan, as a diagnostic agent for hypothalamo-pituitary function, developed by Kaken as KP-102 D. The 2004 development monograph in Drugs R&D (2004;5(4):236–9) records it awaiting that approval with a launch planned for that year, alongside separate Japanese phase 2 work for short stature under the designation KP-102 LN.
The same monograph records what happened in the United States: Wyeth held the North American sublicense and ran phase 2 trials in growth hormone deficiency, and development appears to have been discontinued. Drugs@FDA and DailyMed both returned nothing for pralmorelin on 2026-09-13.
Why the diagnostic approval argues against the sales pitch
The reasoning behind the test is worth setting out, because it is the most useful thing in this compound's file and it is almost always reported backwards.
The test works because the injection raises growth hormone markedly in healthy people, irrespective of sex, obesity or age, while in people whose pituitary is failing the rise is significantly smaller. A published receiver-operating-characteristic analysis set the cut-off at a peak growth hormone of 15.0 µg/L to separate growth hormone deficiency from health.
That is a single provoked measurement, designed to be unambiguous. A regulator satisfied that the spike is reliable enough to diagnose a disease has certified exactly one property: the pituitary responds. It has certified nothing about what happens if a person provokes that response every day for six months, which is the question the market is actually asking. The approval that sounds like vindication is the approval of a thermometer, not of a medicine.
The 43 randomized trials, and what they were testing
260 PubMed records name this compound or its synonyms as of 2026-09-13, and 43 carry the randomized-controlled-trial publication type. On its face that is one of the better human files in this library.
Reading the titles changes the picture. They are concentrated in endocrine physiology:
- Secretagogue type, sex-steroid milieu, and abdominal visceral adiposity individually determine secretion — European Journal of Endocrinology, 2010
- Short-term testosterone supplementation relieves growth hormone autonegative feedback in men — Journal of Clinical Endocrinology and Metabolism, 2004
- Differential pulsatile secretagogue control of GH secretion in healthy men — American Journal of Physiology, 2013
- Estradiol regulates GH-releasing peptide's interactions with GH-releasing hormone and somatostatin — European Journal of Endocrinology, 2014
- Determinants of GH-releasing hormone and GH-releasing peptide synergy in men — American Journal of Physiology, 2009
In every one of these the peptide is injected so that a growth hormone pulse can be generated and then studied. The question under test is how estradiol, or testosterone, or visceral fat, or somatostatin changes the pituitary's response. The compound is the instrument; the subject is the axis. A randomized design is used because that is the right way to run a crossover physiology experiment, not because anyone was assessing a treatment.
One more detail belongs on the record. At least one of the 43 is not a human trial at all: a 2016 PLoS One paper tests GHRP-2 and cysteamine administration on growth performance in livestock. A count of randomized controlled trials includes it, because the publication type describes the design, not the species.
And the earliest human papers, from 1995 in Metabolism and Clinical Endocrinology, are exactly what they appear to be — first-in-human demonstrations that an intravenous bolus releases growth hormone. Real, replicated, and about a hormone level.
What is not in the record
No trial in the published literature uses muscle mass, strength, body composition, injury recovery or sleep quality as an endpoint. This is the same hole this library has documented across the whole secretagogue family, and it is not a reporting gap: the hormone response is the most studied thing about these compounds, and the functional consequence has been measured rarely and found small — see sermorelin, ipamorelin and CJC-1295 for the same pattern with different molecules.
ClinicalTrials.gov held zero registrations on 2026-09-13, searched under both GHRP-2 and pralmorelin. That emptiness is chronological rather than meaningful: the human work predates registration as a publishing condition. It is the same reason GHRP-6's registry page is blank, and it is a different situation from a compound nobody has ever dosed.
In sport it is a prohibited growth hormone secretagogue, and a 2010 paper in Rapid Communications in Mass Spectrometry sets out how to detect it and its metabolite in urine — a reminder that for some compounds, part of the literature exists so that laboratories can find them rather than because anyone is studying a benefit.
How to read this entry
GHRP-2 is better documented than most of this library and better documented than its reputation suggests — in one narrow direction. That it raises growth hormone in a person is not in doubt; a regulator has built a diagnostic test on it. What has never been shown is that doing so repeatedly changes anything about a body.
For what this class of compound is in general, see what peptides are; for the compound most often sold beside it, GHRP-6.