Impurities in compounded GLP-1s: what has actually been measured
Two laboratories published measurements of compounded GLP-1 products in 2026, and both found impurity profiles the approved drugs don't carry. One of those labs belongs to Novo Nordisk. The other belongs to Eli Lilly. That doesn't make the measurements wrong, and it's the first thing anybody should say about them — because nobody else has published a comparable analysis, and the reason is that nobody else has the reference material.
What this comes down to
- A July 2026 analysis found compounded and follow-on semaglutide carrying amino acid deletions and additions the originator doesn't have, plus unidentified impurities, and showed those impurities presenting potentially immunogenic peptides in a cell assay.
- A May 2026 analysis identified a previously unknown impurity in compounded tirzepatide combined with vitamin B12, formed by a reaction between the two. The authors say the clinical effects are unknown.
- Not one randomized controlled trial of a compounded semaglutide or tirzepatide product exists in PubMed. The same filter returns 413 for the molecules themselves.
- The FDA says the salt forms some compounders use are different active ingredients, and that it is not aware of any lawful basis for using them.
What the two laboratories measured
The larger of the two studies ran multiple samples of follow-on and compounded semaglutide and liraglutide through four methods: a major histocompatibility complex-II peptide proteomics assay, liquid chromatography-mass spectrometry, photostability testing and a fibrillation assay[1]. The compounded and follow-on products carried distinct impurity profiles against the originators — amino acid deletions and additions, plus impurities the authors could not identify — and when those impurities were used to stimulate dendritic cells from healthy donors, potentially immunogenic peptides were presented in numbers and distributions the originator products did not produce.
The light-exposure arm is the part with the most practical bite. Exposed to light, compounded semaglutide and the originator products diverged significantly in strength, in total impurity and in high-molecular-weight protein. A vial that spent a summer afternoon on a doorstep is not a hypothetical on these platforms, which is why how these products are stored and shipped is a separate question worth its own page.
The second study is narrower and, for this site's readers, closer to home. Samples of compounded tirzepatide combined with analogs of vitamin B12 were obtained from various sources in the U.S. market and tested for peptide-related impurities. The authors identified a widespread, previously unidentified impurity produced by a chemical reaction between tirzepatide and certain B12 analogs[2]. Their own conclusion says the clinical effects of that impurity are unknown.
That combination is not rare here. Twenty priced lines across nine of our 242 sellers are a GLP-1 mixed with B12, and we cover why sellers build it that way in the B12 combination article. Two of them are StackMD and Balanced Hormone Health. A separate 2026 survey of 33 compounded semaglutide and tirzepatide products found 48% of them combining the peptide with some mix of cyanocobalamin, glycine, niacinamide, docusate or ondansetron[6], so the ingredient that produced the impurity is one of the commonest additions in the category.
Every author of the semaglutide paper is an employee or shareholder of Novo Nordisk. Every author of the tirzepatide paper works for Eli Lilly. Both companies sell the approved products these compounded versions compete with, and both disclosed it plainly in the papers. Treat the findings as measurements that need independent replication — and notice that the independent replication does not exist.
PubMed, via the E-utilities API
("compounded semaglutide"[tiab] OR "compounded tirzepatide"[tiab]) AND (impurit*[tiab] OR purity[tiab] OR contaminat*[tiab])Returned 2. Positive control — heparin, identical filter — returned 962 through the identical filter in the same session.
Heparin is the control on purpose: it is the drug whose contamination crisis built the modern rules for testing an imported active ingredient, and the same filter finds nearly a thousand papers on it. Two is the entire published record on compounded GLP-1 impurities, and both of the two are named above.
Nobody has run the trial
The question a reader actually wants answered is whether a compounded vial works and behaves like the approved pen. There's no trial. Filtered to randomized controlled trials in humans, PubMed returns nothing for compounded semaglutide, compounded tirzepatide or compounded GLP-1 as a phrase. The filter works: the same one returns 413 randomized trials of the molecules themselves.
PubMed, via the E-utilities API
("compounded semaglutide"[tiab] OR "compounded tirzepatide"[tiab] OR "compounded GLP-1"[tiab]) AND "randomized controlled trial"[pt] AND "humans"[mh]Returned 0. Positive control — semaglutide or tirzepatide, identical filter — returned 413 through the identical filter in the same session.
The control is the same filter over the same molecules, which is what makes the zero mean something. Compounded products are not required to be trialed before they are sold — that is the point of the compounding exemption, not a failure of anyone's diligence — so the absence is expected. It is also the reason the efficacy claims on these pages rest on the approved products' trials rather than on the product being shipped.
What the FDA says it has seen
The agency keeps a single page for its concerns with unapproved GLP-1 drugs, and the version we read carries a face date of September 1, 2026[3]. It is worth reading in full; these are the parts that bear on what's in the vial.
- Salt forms. Semaglutide sodium and semaglutide acetate are different active ingredients from the one in the approved drugs, and the agency says it is not aware of any lawful basis for using them in compounding.
- A border action. The FDA has established a green list import alert, numbered 66-80, aimed at stopping GLP-1 active ingredients with potential quality concerns from entering the U.S. supply chain, while leaving imports from manufacturers that appear compliant untouched.
- Fraudulent labels. The agency says it is aware of compounded semaglutide and tirzepatide carrying false label information, including labels naming compounding pharmacies that do not exist, and labels naming a real pharmacy that did not make the product.
- The reporting counts. As of May 31, 2026 the agency had received 990 adverse event reports associated with compounded semaglutide and more than 730 associated with compounded tirzepatide.
“However, federal law does not require state-licensed pharmacies that are not outsourcing facilities to submit adverse events to FDA so it is likely that adverse events from compounded versions of these drugs are underreported.”
[3]
Underreporting means the counts above are floors, not totals. It also means nobody can turn them into a rate, because the denominator — how many people took a compounded GLP-1 — is not collected either. A number with no denominator is a signal to look harder, not a risk you can compare against anything.
The adverse-event database, read carefully
One independent group has gone through the FDA's adverse event reporting system for this specifically, covering 2018 to 2024. Of 81,078 GLP-1 reports, 707 involved compounded products. Against non-compounded formulations, compounded products showed higher reporting odds for contamination, at 19.00, for compounding or manufacturing issues, at 8.51, for preparation errors, at 48.92, and for hospitalization, at 2.35[4]. They also showed lower reporting odds for administration errors, at 0.29, and for dosing errors, at 0.24.
It compares how often a term shows up in reports about one product against how often it shows up in reports about another. Who files a report, and what they think is worth mentioning, differs between a prefilled pen from a pharmacy counter and a vial from a telehealth platform. The lower dosing-error odds in that same analysis sit beside a poison control series describing tenfold errors with compounded vials, which is a contradiction best explained by who bothers to file.
The part a buyer can check
None of the above is checkable from a seller's homepage. What is checkable is whether the company names the pharmacy that compounds its drug, whether the product arrives with instructions for use, and whether the label's pharmacy name and address survive a search — all three are on the FDA's own list of telehealth warning signs[3]. Whether a company publishes that much is one of the things our grades measure, and every company we've read is in the roster.
Compounded is not a synonym for counterfeit, and it isn't a synonym for equivalent either. The agency's own summary of the difference is short: compounded drugs are not FDA approved, which means it does not review them for safety, effectiveness or quality before they are marketed[5]. What that means legally, and what a 503A pharmacy is allowed to do, is set out in compounded versus brand; where the FDA's position now stands, with dates, is in the compounding timeline.
Questions people actually ask
Are compounded GLP-1s tested for impurities before they are sold?
Not by the FDA. Compounded drugs are not FDA approved, which means the agency does not review them for safety, effectiveness or quality before they reach a patient. A 503A pharmacy operates under state licensure and pharmacy standards; an outsourcing facility registered under 503B is subject to more federal oversight. Neither is a premarket review of the finished product.
What did the 2026 impurity studies actually find?
Distinct impurity profiles against the approved products — amino acid deletions and additions plus unidentified impurities in compounded and follow-on semaglutide, and a previously unknown impurity in compounded tirzepatide combined with B12 analogs, formed by a reaction between the two ingredients. Both papers were written by employees of the companies that make the approved drugs.
Does an impurity mean the drug will hurt me?
The papers do not say that, and neither do we. The semaglutide study reports increased immunogenicity potential in a laboratory assay, not harm in people. The tirzepatide study says the clinical effects of the impurity it found are unknown. What is missing is the human evidence either way, because no randomized trial of a compounded GLP-1 product has been published.
Is semaglutide sodium the same as semaglutide?
The FDA says no. Its position, on the page we read dated September 1, 2026, is that salt forms including semaglutide sodium and semaglutide acetate are different active ingredients from the one used in the approved drugs, that the agency has no information on whether they share the same chemical and pharmacologic properties, and that it is not aware of any lawful basis for their use in compounding.
Sources
Every source here was fetched and read for this article, with the identifier taken off the record that came back and the claim it supports written down beside it. All of it was read in September 2026, the same session the rest of this page draws on.
- 1.Impurities and Potential Immunogenicity Associated With Follow-on and Compounded Glucagon-like Peptide-1 Receptor Agonists. Pharmaceutical Research, 2026. PMID 42533250 · doi:10.1007/s11095-026-04146-9Impurity profiles of follow-on and compounded semaglutide and liraglutide measured by MHC-II peptide proteomics, LC-MS, photostability testing and a fibrillation assay; amino acid deletions and additions plus unidentified impurities against the originators; potentially immunogenic peptides presented on impurity-stimulated dendritic cells; significant disparity in strength, impurity sum and high-molecular-weight protein on light exposure; and the authors' declaration that they are employees and shareholders of Novo Nordisk.
- 2.A novel, widespread impurity in mass-compounded tirzepatide/B12 products: potential patient safety implications. Expert Opinion on Drug Safety, 2026. PMID 42010938 · doi:10.1080/14740338.2026.2663185A widespread, previously unidentified impurity in compounded tirzepatide combined with B12 analogs, produced by a chemical reaction between the two; the authors' statement that the clinical effects of the impurity are unknown; and their affiliation with Eli Lilly and Company.
- 3.FDA's Concerns with Unapproved GLP-1 Drugs Used for Weight Loss. U.S. Food and Drug Administration, 2026. Source · Document dated September 2026The salt-forms position on semaglutide sodium and semaglutide acetate; import alert 66-80; fraudulent compounded products carrying labels of pharmacies that do not exist; 990 adverse event reports for compounded semaglutide and more than 730 for compounded tirzepatide as of May 31, 2026; the verbatim sentence on underreporting; and the telehealth warning signs list.
- 4.Safety analysis of compounded GLP-1 receptor agonists: a pharmacovigilance study using the FDA adverse event reporting system. Expert Opinion on Drug Safety, 2026. PMID 40285721 · doi:10.1080/14740338.2025.2499670Of 81,078 GLP-1 reports in FAERS from 2018 to 2024, 707 involved compounded products, with adjusted reporting odds ratios of 19.00 for contamination, 8.51 for compounding or manufacturing issues, 48.92 for preparation errors, 2.35 for hospitalization, 0.29 for administration errors and 0.24 for dosing errors.
- 5.Compounding and the FDA: Questions and Answers. U.S. Food and Drug Administration, 2025. Source · Document dated September 2025Compounded drugs are not FDA approved, meaning the agency does not verify their safety, effectiveness or quality before they are marketed, and the distinction between a 503A pharmacy and a registered outsourcing facility.
- 6.Compounded Semaglutide and Tirzepatide Products Use Unique Formulations but Efficacy and Safety Largely Unknown. Annals of Pharmacotherapy, 2026. PMID 41689811 · doi:10.1177/10600280261421979Thirty-three unique compounded semaglutide or tirzepatide products identified on compounding sites between February 3 and March 20, 2025, of which 48% combined the peptide with cyanocobalamin, glycine, niacinamide, docusate or ondansetron.
Key figures