The trial evidence for sermorelin is nineteen people over five months, and its only reported adverse effect was a transient rise in blood lipids that resolved by the end. That is a thin basis for a safety opinion, and it is most of what exists. There is one other place to look: the FDA’s adverse event reporting system, where anyone — a doctor, a manufacturer, a patient — can file a report about a drug.
Fifty-nine of those reports name sermorelin. They are worth reading and they are worth reading carefully, because the system is built to catch signals rather than to measure risk, and almost every intuitive way to use the number is wrong.
What the reports contain
Queried on September 15, 2026, 59 reports name sermorelin, spread from 2009 to 2026 and clustering in two years: fifteen in 2016 and thirteen in 2025 [1]. Thirty-six are classified serious and twenty-three are not.
The reactions cluster tightly around one theme. Itching leads at six reports, then nausea at five, then hypersensitivity, malaise and rash at four each, then anaphylactic reaction, burning sensation, redness, sweating, injection-site itching and injection-site hives at three each [1]. That is an allergic and injection-site picture, which is what you would expect from an injected peptide and is not what the one trial reported. Three reports carry the term “wrong product administered,” which is its own kind of signal for a drug that is compounded rather than manufactured to a reviewed label.
Why the number cannot be turned into a risk
Three things stop it, and all three are properties of the system rather than of sermorelin.
There is no denominator. Nobody knows how many people have taken sermorelin, so fifty-nine reports cannot become a rate. It could be fifty-nine out of a thousand or out of a million, and the database has no opinion on which.
A report is not a finding. Filing one requires only that somebody suspected a connection. Nothing is verified, nothing is adjudicated, and duplicate reports of the same event exist.
And the reports are rarely about one drug. This is the one that matters most here. The median report names seven products, and the largest names fifty-three. Only nine of the fifty-nine name sermorelin on its own [1]. The drugs appearing most often alongside it are Xyrem, testosterone cypionate, Adderall, vitamin D and fish oil — which describes the kind of patient who ends up on sermorelin more than it describes sermorelin.
The death report, and why it proves nothing
One report among the fifty-nine is flagged as a death. It is the sort of fact that gets quoted on its own, so here is the whole of it: that report names seven products, among them testosterone cypionate and the bacteriostatic water used to mix an injection [1].
“A death has been reported with sermorelin” is a true sentence and a misleading one. The report establishes that somebody died and that sermorelin was one of seven things named. It does not establish which, or any, of those seven was responsible, and the reporting system is not designed to answer that. Anyone citing it as evidence sermorelin is dangerous, or omitting it as evidence sermorelin is safe, is using it for something it cannot do.
A counting problem worth knowing about
The database records the drug under at least six different names. “Sermorelin” appears twenty-five times, “sermorelin acetate” seventeen, and “sermorelin acetate” with a trailing period thirteen more — a full stop is enough to make a separate database term [1]. Ten further entries describe combination kits pairing it with GHRP-2 and GHRP-6, one of them naming the compounding pharmacy in the product itself.
So anyone searching a single spelling sees fewer than half the reports and has no way to know it. That is the same naming split the acetate question runs into from the other direction, where the scientific literature uses one name and the pharmacy shelf uses the other.
What this is actually good for
Not risk, and not reassurance. What a reporting system is good at is telling you which kinds of problem have come up at all, so you know what to watch for and what to mention. Here that is allergic and injection-site reactions — itching, rash, hives at the site, swelling, and in three reports anaphylaxis — plus nausea and a general feeling of being unwell.
None of those appears in the adult trial, which measured a narrower set of things in nineteen people and would have struggled to catch an uncommon reaction at that size. The two sources are not in conflict; they are looking at different questions. A trial asks what happens on average in a controlled group. A reporting system asks what has ever gone wrong badly enough that somebody wrote it down.
If you are starting sermorelin, the practical version is short: an injected peptide can cause an allergic reaction, a handful of reports describe severe ones, and that is worth knowing before the first dose rather than during it. What to ask a prescriber covers where that conversation belongs.