Study finds hair-loss signals for semaglutide and tirzepatide in FDA reports

A ten-year review of reports to the FDA's adverse-event database found statistical signals for hair loss with semaglutide and tirzepatide, researchers reported on June 7, 2026, in the Journal of Cosmetic Dermatology ^[1]. The same analysis found no such signal for five other GLP-1 receptor agonists ^[1].

What the study looked at

The authors, led by Aditya K. Gupta of the University of Toronto, examined the FDA Adverse Event Reporting System (FAERS) from the first quarter of 2016 through the third quarter of 2025 ^[1]. They ran disproportionality analyses, a method that flags drug-and-event pairs reported more often than expected relative to everything else in a database, for seven GLP-1 receptor agonists: albiglutide, dulaglutide, exenatide, liraglutide, lixisenatide, semaglutide and tirzepatide ^[1]. They searched for seven alopecia-related terms, including alopecia, alopecia areata, alopecia totalis, androgenetic alopecia and diffuse alopecia ^[1].

What they found

The study counted 1,941 reports across 1,929 cases of alopecia-related events involving GLP-1 drugs from 2016 to 2025 ^[1]. Semaglutide (sold as Ozempic, Wegovy and Rybelsus) accounted for the largest number, 697 reports ^[1].

Signals, defined in the study as a reporting odds ratio whose lower 95% confidence bound sits above 1, appeared for semaglutide in 2022 (reporting odds ratio 1.43, 95% confidence interval 1.15 to 1.77) and again in 2024 (1.55, 95% CI 1.35 to 1.77) ^[1]. The researchers also reported signals for tirzepatide (Mounjaro, Zepbound) in 2024 and 2025 ^[1]. No signals were found for albiglutide, dulaglutide, exenatide, liraglutide or lixisenatide ^[1].

Among semaglutide reports that contained follow-up information, 25 percent documented a positive dechallenge, meaning the hair loss improved or resolved after the drug was stopped, and 18 percent documented a positive rechallenge, meaning it returned after the drug was restarted ^[1]. Those proportions were calculated only among reports that had dechallenge or rechallenge data, not among all 697 semaglutide reports ^[1].

What the findings can and cannot show

FAERS holds voluntarily submitted reports. The database cannot be used to calculate how often a side effect occurs in people taking a drug, and it does not let researchers adjust for other reasons a person might lose hair, such as illness, stress or other medications ^[1]. Reports with missing dechallenge or rechallenge values were excluded from those calculations rather than filled in ^[1].

A reporting signal means a drug-and-event pair shows up more often in reports than expected. It is not proof that the drug caused the event, and the study does not provide a risk figure for any product.

Hair loss entered the conversation earlier: the FDA flagged alopecia as a potential effect of GLP-1 receptor agonists in July 2023, and alopecia was listed among adverse events in the STEP 3 randomized trial of semaglutide published in 2021 ^[1].

Why it matters for patients

The study gives a shape to a complaint that has circulated among people using GLP-1 drugs, but it does not produce a risk number. It does not say how many people take these drugs, how many develop hair loss, or whether the drugs are responsible ^[1]. For someone taking semaglutide or tirzepatide, the results cannot establish that the medication is causing any hair thinning, and the paper does not call for a change in how the drugs are prescribed.

The paper notes that alopecia can affect mental health, which is part of why a visible side effect draws attention even when it is not dangerous ^[1]. The practical value here is that the pattern appears in the FDA's own reporting system, the tool regulators use to decide whether a safety question needs a closer look. What a patient does with that information, including whether to raise hair loss with a prescriber, is a decision the study itself does not address.

What happens next

The authors write that their findings merit further monitoring of GLP-1 receptor agonists for hair loss ^[1]. The paper describes no regulatory action or label change, and no follow-up study has been announced ^[1]. Whether the FDA will revisit the question has not been disclosed.

No published trial data has established how often alopecia occurs with these drugs, and the study's authors say FAERS data cannot answer that question ^[1].