Sermorelin Side Effects: What the Trials, the Class Label, and the Cancer Data Actually Report

Roughly 9,800 people a month search "sermorelin side effects," and a surprising number of them add one more word: cancer. That second query is the one this article was written for, because it is a real, specific, answerable question about a drug that works by raising growth hormone — and almost every page ranking for it either waves it away in a sentence or leans on a scary-sounding mechanism with nothing behind it. Both are avoidable. There is real evidence here, and it points somewhere specific.

Before any of it, one fact that reframes everything below, and that we verified rather than assumed: sermorelin has no current FDA-approved drug label in the United States. Its approved brand, Geref, holds two applications in FDA's own database — NDA 019863 for a 0.05 mg ampule and NDA 020443 for 0.5 mg and 1 mg vials — and every product under both is listed with a marketing status of "Discontinued," each carrying FDA's own note that the withdrawal was not for safety or effectiveness reasons⁴. A direct search of DailyMed, the National Library of Medicine's official archive of FDA-approved drug labeling, returns zero results for sermorelin against a database published July 31, 2026⁵. So there is no "Adverse Reactions, section 6" to quote for this drug. Anyone who hands you a tidy, official-sounding sermorelin side-effect table is not reading it off a label, because there isn't one to read.

What there is: two published trials that recorded tolerability directly, a small spontaneous-report database, and — the most useful and least-cited source — a current FDA label for the one growth-hormone-releasing-hormone analog that is still marketed, which carries warnings sermorelin's own discontinued label predates.

What sermorelin's own studies actually reported

The clinical literature on sermorelin is thin, old, and mostly pediatric, and its tolerability findings are correspondingly narrow. The compound's own clinical review — covering both its diagnostic use (a single 1 µg/kg intravenous dose to provoke a GH pulse) and its treatment use in prepubertal children with idiopathic growth hormone deficiency at 30 µg/kg nightly — describes both single intravenous and repeated daily subcutaneous dosing as "well tolerated," and names exactly two common adverse events: transient facial flushing and pain at the injection site¹. That is the entire commonly-reported adverse-event list from sermorelin's own approval-era literature.

The best-controlled adult data comes from a 1997 randomized, placebo-controlled trial in 19 adults aged 55 to 71 — nightly subcutaneous self-injection of a closely related, stabilized GHRH(1-29) analog at 10 µg/kg, four weeks of placebo followed by sixteen weeks of active drug, with safety labs (complete blood count and chemistry panel) drawn at baseline, after the placebo month, and at 2, 4, 8, 12, and 16 weeks. Its safety conclusion is a single sentence: "The only adverse side-effect was transient hyperlipidemia, which resolved by the end of the study²." Blood pressure and body weight were unaffected; fasting insulin and glucose were unaltered². One precision this site keeps stating rather than quietly dropping: the molecule in that trial is [Nle27]GHRH(1-29)-NH₂, the same 29-residue backbone with one oxidation-resistant substitution, not byte-for-byte the sermorelin sold today — the closest controlled adult safety data that exists for this drug class, not a sermorelin trial in the strict sense.

The one real-world adult dataset — a retrospective review of 105 hypogonadal men on testosterone therapy, 14 of whom met strict compliance criteria — reports a significant IGF-1 rise over a mean 134 days but studied sermorelin combined with two other secretagogues (GHRP-2 and GHRP-6), so it cannot attribute an adverse event to sermorelin specifically any more than it can attribute a benefit³.

The post-marketing picture: 58 reports, and why that number matters both ways

FDA's Adverse Event Reporting System is the closest thing to a real-world safety signal for a drug with no current label. Counted live against openFDA on August 3, 2026, exactly 58 reports in the entire database name sermorelin among the patient's drugs. The most frequently coded reactions across them: pruritus (6 reports), nausea (5), hypersensitivity, malaise and rash (4 each), and anaphylactic reaction, injection-site pruritus, injection-site urticaria, erythema, hyperhidrosis and swelling (3 each)⁶.

Read that in both directions, because both readings are honest. On one hand, the pattern is unremarkable and matches the trial literature: local injection-site and hypersensitivity-type reactions, with some gastrointestinal upset. On the other, 58 spontaneous reports across a database of millions is far too few to compute an incidence rate from, and adverse-event reporting for compounded drugs is structurally weak — federal law does not require state-licensed pharmacies that are not outsourcing facilities to submit adverse events to FDA at all. A small number in FAERS is not evidence of safety. It is evidence of thin reporting.

The class label nobody quotes — and it is the one with the cancer warning

Here is the source that actually answers the question people are searching. Sermorelin has no current label, but tesamorelin does — marketed as EGRIFTA SV, it is a GHRH analog working through the same pituitary receptor sermorelin does, and it is the only FDA-approved drug of that class currently on the U.S. market. Its label is live, current, and specific.

Section 5.1 of that label is titled, in FDA's own words, "Increased Risk of Neoplasms." Its reasoning is stated in one sentence that applies mechanistically to every GHRH analog, sermorelin included: the drug "induces the release of endogenous growth hormone (GH), a known growth factor." The label's consequences are concrete rather than hedged — patients with active malignancy are not to be treated with it at all; patients with a history of treated and stable malignancies are to be started only after careful evaluation of benefit against "the risk of re-activation of the underlying malignancy"; and treatment is to be discontinued "if there is any evidence of recurrent malignancy⁷."

Section 5.2, "Elevated IGF-1 Levels," is the quieter and arguably more relevant one for a healthy adult buying a GH secretagogue online. It states plainly that "the effects of prolonged elevations in IGF-1 levels are unknown," instructs prescribers to monitor IGF-1 during therapy, and to consider discontinuing in patients with persistent elevations above 3 standard deviation scores. It also quantifies how often that actually happens: among patients on 26 weeks of treatment, 47% had IGF-1 above 2 SDS and 36% above 3 SDS, with the effect visible as early as week 13⁷. A raised IGF-1 is not a side effect of a GHRH analog — it is the intended pharmacology, which is exactly why "monitor it, and stop if it stays too high" is a labelled instruction rather than a footnote.

That is a real answer to "sermorelin side effects cancer," and it is not a comfortable one in either direction: the closest labelled relative of sermorelin carries an explicit neoplasm warning and an explicit contraindication in active malignancy, and simultaneously the label does not claim the drug causes cancer — it says the long-term consequences of sustained IGF-1 elevation are not known.

What the epidemiology says about the IGF-1 → cancer chain

The mechanistic worry has a real evidence base, and it is more modest than the fear suggests. A systematic review and meta-regression of 21 studies (3,609 cancer cases, 7,137 controls) found high circulating IGF-1 associated with increased risk of prostate cancer (odds ratio 1.49, 95% CI 1.14–1.95, comparing the 75th with the 25th percentile) and premenopausal breast cancer (1.65, 1.26–2.08), while concluding in its own words that "associations are modest and vary between sites⁸." That is an association between naturally-occurring IGF-1 levels and cancer risk — not a finding about a drug.

The closest thing to a direct test is decades of growth hormone therapy itself, which raises IGF-1 far more aggressively than a secretagogue does. The SAGhE European cohort followed 23,984 patients treated with recombinant human GH across eight countries since 1984. Its conclusion: "Our results do not generally support a carcinogenic effect of r-hGH." The caveats it reports alongside that are worth carrying rather than dropping — raised risks were "largely consequent on second primary malignancies in patients given r-hGH after cancer treatment," with "no clear raised risk in patients with growth failure without other major disease"; only bone and bladder cancer incidence was significantly raised in previously cancer-free patients; cancer risk was unrelated to duration or cumulative GH dose; but among patients treated after a previous cancer, cancer mortality rose significantly with increasing daily dose⁹. A separate pair of observational studies covering 37,702 patients and 130,476 patient-years found "no indication of increased mortality risk nor [adverse event] incidence related to GH dose in any risk group¹¹."

And on the specific question of GHRH analogs rather than GH itself, the honest answer is that the preclinical data contradicts itself. A 2019 review by the laboratory that has studied these molecules longest reports both halves without resolving them: in vitro, GHRH agonists "stimulate growth of human cancer cells and upregulate" the GHRH receptor; in vivo, the same agonists "inhibit growth of human cancers xenografted into nude mice" and downregulate both pituitary and tumoral GHRH receptors¹⁰. Anyone citing only the first clause is scaremongering; anyone citing only the second is selling. Both are in the same paper.

Two labelled effects that are not cancer and are far more likely

Because the cancer question dominates the search box, two much more probable effects get skipped. The tesamorelin label reports both, from a 543-patient placebo-controlled phase.

Glucose intolerance. Elevated HbA1c (≥6.5%) from baseline to week 26 occurred in 5% of treated patients versus 1% on placebo, an intent-to-treat hazard ratio of 3.3 (95% CI 1.4–9.6) for developing diabetes. The label instructs prescribers to evaluate glucose status before starting and monitor periodically thereafter⁷. Raising growth hormone raises insulin resistance; that is basic endocrinology, and it is the effect most likely to actually show up on a blood panel.

Fluid retention. The label attributes this directly "to the induction of GH secretion," manifesting as edema, joint pain, and carpal tunnel syndrome, which are "either transient or resolve with discontinuation⁷." In the 26-week table, injection site reactions ran 17% versus 6% on placebo, arthralgia 13% versus 11%, and peripheral edema 6% versus 2%⁷.

Again: that is tesamorelin's data, not sermorelin's. Different molecule, different dose, a population of HIV-positive patients with lipodystrophy rather than midlife adults chasing body composition. It is offered as the labelled behaviour of this drug class — the best-documented thing available — not as sermorelin's own numbers, which do not exist at label-grade quality.

Where that leaves a reader

Sermorelin's own recorded adverse events are mild and local: injection-site pain, transient flushing, a transient lipid rise in the one controlled adult trial. Its post-marketing file is 58 reports of mostly local and hypersensitivity-type reactions. The realistic concerns are not exotic — they are glucose tolerance and fluid retention, both labelled for its closest marketed relative, both checkable on ordinary bloodwork. And the cancer question has a real answer with a real shape: the class label warns about neoplasms and forbids use in active malignancy; large GH cohorts do not show a general carcinogenic effect in people without a prior cancer; and the long-term consequence of sustained IGF-1 elevation is stated on an FDA label as unknown, which is not the same as safe.

None of that is a reason to skip the more mundane question this site exists to answer, which is who you would actually get it from. Sermorelin is prescription-only and, with Geref discontinued, every sermorelin sold in the United States today is a compounded product — which means the pharmacy, its regulatory category, and what the seller does and does not disclose are not side details, they are the product. Our sermorelin provider rankings read those disclosures off each company's own site, and several of them get it wrong in ways worth seeing before you buy — MangoRx's own review, for one, covers a provider with a real, flat, "price locked in" sermorelin rate but a pharmacy described only by accreditation, never by name. For what the drug is actually shown to do on a timeline, our walk through the sermorelin before-and-after evidence separates the measured findings from the anecdotal ones, and our head-to-head on sermorelin, ipamorelin and CJC-1295 covers how the three compare on mechanism and regulatory status. Once a prescriber has given you an actual dose in milligrams, our sermorelin reconstitution calculator converts it into the syringe volume that dose corresponds to — arithmetic on a number you were given, never a suggestion of what that number should be.