When Fewer Biologic Starts Look Like Less Disease, Check Who Got the Drug First

A paper posted online September 12 in the Journal of the American Academy of Dermatology, by Rahib K. Islam, Lacey M. Falgout, Nicholas J. Culotta, Christopher J. Sayed, and Christopher J. Haas, reports that patients with hidradenitis suppurativa who were exposed to GLP-1 receptor agonists were less likely to subsequently start a biologic than patients who were not. The finding fits a growing pile of retrospective data suggesting these drugs, originally built for type 2 diabetes and obesity, do something to HS beyond weight loss. The authors frame their central question directly in the title: does the reduced biologic initiation reflect a real drop in disease severity, or does it reflect who gets offered a GLP-1 prescription in the first place.

That framing matters because the underlying data are observational, drawn from patient records rather than a randomized trial, so anyone who received a GLP-1 receptor agonist already differed from anyone who did not in ways that are hard to fully capture. A separate cohort study indexed on PubMed (record 42384412) examined mortality, cardiovascular events, and systemic complications in HS patients on GLP-1 therapy, work that shares the same structural limitation of comparing groups that were never assigned to treatment by chance.

Access itself has been quantified elsewhere. A study in PMC (article PMC12861177) looking at psoriasis and HS patients found that among those who met eligibility criteria for GLP-1RA therapy, only 36.4% received counseling about the drug class and just 22.1% were actually prescribed one. Among 566 eligible but uncounseled patients, 18%, or 102 people, lacked a primary care physician at all, and of those, nearly a quarter saw only a dermatologist and one other specialist. That pattern points toward a plausible confound for the new JAAD analysis: patients who obtain a GLP-1 prescription may be systematically better connected to care, more insured, or more engaged with a physician network, any of which could independently lower their odds of escalating to a biologic.

Mechanistic claims add a layer of plausibility without resolving the access question. Research indexed under PubMed record 42389771 describes GLP-1 receptor agonists inhibiting nuclear factor kappa-light-chain-enhancer of activated B cells, a pathway tied to inflammation, and suggests these drugs may enhance the efficacy of biologics used for HS. But a systematic review in the American Journal of Clinical Dermatology (published by Springer Nature) and a companion review in PMC (article PMC13116881) both stop short of confirming a dose-dependent anti-inflammatory signal outright, noting that inflammatory parameters showed no significant changes in some analyses even as clinical scores improved and higher-dose patients trended toward better outcomes.

A retrospective French cohort covered by AJMC, drawing on results published in JAMA Dermatology, offers a concrete efficacy benchmark rather than an access argument: at six months, 54% of patients on semaglutide had at least a one-point reduction in HS-PGA score, and 12% had a two-point or greater reduction. AJMC's own writeup cautions that the retrospective design, plus the fact that some patients remained on other HS medications concurrently, prevents definitive conclusions about semaglutide's isolated effect.

The JAAD authors' contribution is not new outcome data so much as a framing exercise applied to an existing signal, asking readers to hold two explanations open at once rather than defaulting to the more flattering one. A drop in biologic starts after GLP-1 exposure could mean the drug is doing real anti-inflammatory work, consistent with the JAMA Dermatology commentary describing GLP-1RAs as an emerging opportunity for comorbidity management in HS. It could just as easily mean that the patients who reach a GLP-1 prescription were already less likely to need a biologic, given how unevenly counseling and prescribing were distributed in the PMC access study. Nothing in the current body of retrospective cohorts, PubMed-indexed mortality analyses, or single-country registries can adjudicate between those two readings, because none of them randomized who got the drug.