Compounded GLP-1s: salts, impurities and antibodies

What the FDA reports

The FDA keeps a page on unapproved GLP-1s. "As of May 31, 2026" it had 990 adverse-event reports with compounded semaglutide and more than 730 with compounded tirzepatide, and it notes these are likely underreported¹. It also clarifies a technical point that affects many products: "These salt forms, including semaglutide sodium and semaglutide acetate, are different active ingredients" from the approved drug, and retatrutide and cagrilintide "cannot be used in compounding under federal law"¹.

Pharmacovigilance signals

A FAERS analysis (2018–2024) compared 707 reports for compounded GLP-1s with those for branded products, out of 81,078 GLP-1 reports. Compounded products were associated with more hospitalization (ROR 2.35), product contamination (ROR 19.0) and preparation errors (ROR 48.9)². By contrast, administration and dosing errors were reported proportionally less often with compounded products (ROR 0.29 and 0.24)². It is a disproportionality analysis: it shows signals, not causes or rates.

Case series describe serious dosing errors. One poison center reported three cases with compounded semaglutide, two of them 10-fold dosing errors, and one patient who dosed in milliliters and units rather than milligrams³. Another series described three unintentional semaglutide overdoses at treatment initiation, such as 2 mg instead of 0.1 mg or 2.4 mg instead of 0.25 mg, with nonspecific gastrointestinal symptoms and no hypoglycemia⁴. The FDA itself warned in July 2024 of adverse events, some requiring hospitalization, from dosing errors with compounded injectable semaglutide, citing inexperience with vials and syringes and confusion between milliliters, milligrams and "units"⁹. In a 2026 study, 20 untrained adults relying only on internet information correctly completed 15 of 34 steps when trying to prepare gray-market semaglutide¹⁵.

A new impurity

In 2026, Eli Lilly researchers (the maker of branded tirzepatide) described a previously unidentified, widespread impurity in mass-compounded tirzepatide/B12 products: the result of a chemical reaction between tirzepatide and certain vitamin B12 analogs, present at substantial levels. Its clinical effect is unknown⁵.

The issue almost nobody mentions: immunogenicity

Peptides are immunogenic even when made to pharmaceutical standards. According to the Zepbound label, 64.5% of patients (1,591 of 2,467) developed anti-drug antibodies; 40% cross-reacted with native GIP and 16.5% with native GLP-1. Hypersensitivity was 6.2% in patients with antibodies vs 3% without, and injection-site reactions 11.3% vs 1%. Neutralizing antibodies against activity at the GIP or GLP-1 receptors were uncommon (2.8% and 2.7%), and the label cautions that incidence depends on the assay used⁶. With semaglutide for obesity, the label reports anti-drug antibodies in 3% (50 of 1,709) in the 68-week trials, although it cautions that differences in assay methods preclude comparing incidences across products¹⁰.

Impurities and aggregates can raise that risk: a sequence change can create new T-cell epitopes¹¹. In a case study of generic teriparatide, with FDA co-authors, several impurities were more immunogenic in vitro than the drug itself¹². That is why, in its guidance for synthetic generics of glucagon, liraglutide, nesiritide, teriparatide and teduglutide, the FDA asks for new impurities between 0.10% and 0.5% to be identified, characterized and justified, including comparative immunogenicity testing, and states that a new impurity "> 0.5% is not acceptable"⁷; the guidance was finalized in May 2021¹³. The same logic appears in its risk entry for TB-500: immunogenicity "due to the potential for aggregation as well as peptide-related impurities"⁸.