Two compounds on the growth-hormone shelf are analogs of the same hormone, act at the same receptor, and were both approved by the FDA. One of them is still marketed today. The other has been off the US market since 2008 — and not because anything went wrong with it.
The interesting part is that the two molecules were engineered at opposite ends of the same parent peptide, for reasons that had nothing to do with receptor binding. Sermorelin was made shorter. Tesamorelin was left full-length and given a chemical cap. Comparing them is a clean case study in how a design decision, a regulatory file, and a widely repeated pharmacokinetic number can all come apart from one another.
One parent hormone, two edits
Endogenous growth hormone-releasing hormone circulates as a 44-residue, C-terminally amidated peptide, usually written GHRH(1-44)-NH2 or GRF(1-44)-NH2. Decades of structure-activity work established that the biological information sits in the N-terminal region — the C-terminal tail can largely be removed without abolishing receptor activity.
Sermorelin is the truncation answer. It is GRF(1-29)-NH2: the first 29 residues, C-terminally amidated, commonly reported around 3,358 Da as the free base. It is the shortest fragment that retains substantial GHRH-receptor activity, and it is otherwise chemically identical to the corresponding stretch of the native hormone. Nothing was added.
Tesamorelin is the opposite answer. Per the FDA-approved labeling, it comprises the full 44-amino-acid sequence of human GRF with a hexenoyl moiety — a trans-3-hexenoyl group — attached at the N-terminal tyrosine. The label gives a molecular formula of C221H366N72O67S with a molecular weight of 5,135.9 Da as the free base. The sequence was not shortened at all. A small lipid-like cap was bolted onto position 1.
So one compound removed the end that does not carry the message, and the other protected the end that does.
Why the N-terminus is the position worth defending
The reason tesamorelin's cap sits where it does is enzymology, not receptor chemistry.
GHRH begins Tyr-Ala. A residue of alanine in position 2 is the canonical substrate signature for dipeptidyl peptidase-4 (DPP-4), the exopeptidase that snips the N-terminal dipeptide off susceptible peptides. It is the same enzyme that converts active GLP-1 into its truncated metabolite within minutes, and the same reason semaglutide carries an Aib substitution at that position. We covered the enzyme family in detail in the peptidase and clearance explainer.
Applied to GHRH, DPP-4 cleavage removes Tyr-Ala and destroys the receptor-activating N-terminus. Sermorelin, being native in sequence at positions 1 and 2, is fully exposed to it. The trans-3-hexenoyl group on tesamorelin's Tyr1 physically blocks the enzyme's access to that bond.
This is a real and well-characterised difference, and it is the modification's stated rationale. The trap is in what people assume follows from it.
The half-life claim, checked against the label
Secondary sources routinely report tesamorelin's half-life as roughly 30 minutes, sometimes 26 to 38 minutes, and present it as the payoff for DPP-4 resistance.
The approved labeling says something different. The EGRIFTA SV label states that the mean elimination half-life of tesamorelin was 8 minutes in healthy subjects after a single subcutaneous dose, with a median Tmax of 0.15 hours. The EGRIFTA WR labeling reports a mean elimination half-life of 11 minutes subcutaneously. Both labels also report that absolute bioavailability after subcutaneous administration is less than 4%.
Published human values for sermorelin sit in the same territory — on the order of minutes. Note that the original Geref labeling is no longer carried in the active FDA label databases, so a label-grade sermorelin figure is not available for a like-for-like comparison; treat any single quoted number for it with the usual caution.
The honest read is that blocking DPP-4 did not move tesamorelin's plasma half-life into a different order of magnitude. That is not a failure of the design. It is a reminder that enzymatic degradation is only one clearance route. A ~5 kDa peptide is freely filtered by the glomerulus and captured in the proximal tubule regardless of how protease-resistant it is. Blocking the headline enzyme leaves everything else running — the point we made about neprilysin and renal handling in the same explainer, and the same reason semaglutide needs both an Aib substitution and a fatty diacid.
There is a second, more important reason the number matters less here than it looks. These are secretagogues. The pharmacologically relevant event is the pituitary GH pulse the analog triggers, not the analog's own residence time in plasma. A compound can clear in minutes and still produce a downstream hormonal signal lasting far longer. This is exactly the distinction that separates the GHRH-analog tier from the compounds discussed in CJC-1295 DAC vs No-DAC, where an albumin-binding linker was used to deliberately reshape the duration of receptor exposure rather than merely defend the molecule from an enzyme.
Two regulatory files that answer different questions
This is where the two compounds diverge most sharply, and it is the part vendor copy usually flattens.
Sermorelin was approved twice, for two purposes. Per the FDA's 2013 Federal Register determination, GEREF (sermorelin acetate) injection at 0.05 mg base/amp was approved under NDA 19-863 on December 28, 1990, as a diagnostic agent — for evaluating the ability of pituitary somatotrophs to secrete growth hormone. GEREF at 0.5 mg and 1.0 mg base/vial was approved under NDA 20-443 on September 26, 1997, as a therapeutic for idiopathic growth hormone deficiency in children with growth failure. EMD Serono notified FDA of discontinuation of the diagnostic form on July 11, 2008, and of the therapeutic form on December 2, 2008.
The FDA's determination was that these products "were not withdrawn from sale for reasons of safety or effectiveness." It is worth being precise about what that determination is for: it is the finding that permits abbreviated new drug applications to reference a discontinued listed drug. It is a statement about the regulatory record, not a current endorsement, and not a statement about any use outside the approved indications.
Tesamorelin's file is narrower and still open. It was approved in November 2010 for the reduction of excess abdominal fat in HIV-infected adult patients with lipodystrophy — a single indication it still carries. The pivotal evidence is Falutz et al., N Engl J Med 2007;357(23):2359-2370 (PMID 18057338), a multicentre randomized double-blind placebo-controlled trial in 412 adults with HIV-associated abdominal fat accumulation, in which visceral adipose tissue fell by about 15% in the tesamorelin arm and rose about 5% on placebo over 26 weeks, alongside lipid-profile changes. A second phase 3 trial and extension followed. Theratechnologies has since carried the product through reformulations, with EGRIFTA SV and then EGRIFTA WR reaching the market.
What each compound has actually been tested for
Stating this plainly matters more than the pharmacology.
Sermorelin's human evidence base is built around pediatric growth hormone deficiency and pituitary provocation testing. Those are the questions its trials asked. There is no comparably designed, comparably powered efficacy program establishing outcomes for the adult uses it is most often marketed for.
Tesamorelin's human evidence base is built around visceral adipose tissue in a specific HIV population, and — as we noted in the 2026 GH-axis roundup — its endpoints are imaging and biochemical surrogates. A 2026 meta-analysis in that population found consistent VAT and hepatic-fat reductions with no change in subcutaneous fat or CD4 count. There is no cardiovascular outcomes trial.
So the question "which one has better human data?" is not answerable as stated. They have data about different questions, in different populations, against different endpoints. That is the same reading discipline set out in the evidence hierarchy piece: a large literature about a defined medical use is not general permission to extrapolate.
Documentation notes: what a COA has to account for
The two differ enough analytically that identical paperwork for both is a signal about the template, not the product.
- The theoretical mass must include the modification. Tesamorelin's identity claim is the hexenoylated 44-mer, not bare GRF(1-44). A mass calculated from the unadorned sequence will disagree with the product by the mass of the cap. Same principle as the acetyl and amide corrections in the sequence-reading primer.
- Both are C-terminally amidated. The free-acid impurity differs by about 1 Da from the intended molecule — a near-mass species that a purity percentage does not describe.
- Counterion is not incidental here. The tesamorelin label's own formula is written as the peptide with approximately seven acetate equivalents. Salt form and stoichiometry feed directly into mass accounting; see net peptide content and TFA vs acetate.
- Length sets the plausible route. A 29-mer sits comfortably within solid-phase synthesis; a 44-mer with a synthetic N-terminal acyl group also requires a chemical step, since no ribosome installs a hexenoyl cap. See synthesis routes and impurity profiles.
- Both sequences carry tyrosine but no tryptophan, and both carry methionine — so UV quantitation at 280 nm is workable but weak, and methionine oxidation is a named stability liability.
FAQ
Is tesamorelin just a longer sermorelin? No. The length difference is real but secondary. Sermorelin is native in sequence; tesamorelin carries a synthetic N-terminal acyl group that sermorelin does not have. Those are different kinds of molecule, and only one of them exists in the body.
Does DPP-4 resistance mean tesamorelin lasts longer in circulation? Not to the degree usually claimed. The approved labels report subcutaneous elimination half-lives of 8 to 11 minutes. Protease resistance addresses one clearance route; renal filtration and receptor-mediated processes continue independently.
Why does the FDA determination that sermorelin was "not withdrawn for safety or effectiveness" get quoted so often? Because it is genuine and it is easy to over-read. It confirms that the discontinuation was commercial, and it establishes eligibility for generic applications referencing the discontinued products. It says nothing about uses beyond the approved pediatric-deficiency and diagnostic indications, and nothing about product quality outside that regulatory system.
For where these two sit relative to the ghrelin-mimetic tier and to recombinant GH itself, see the growth-hormone class primer and the GH secretagogue axis explainer. Our documentation standards are described on the quality page, and full compound entries are in the library.
This article is educational and for the laboratory research community. Trulogic Labs products are sold for laboratory and research use only and are not for human consumption.