A Genetics Paper on Acne and IBD Can't Settle the Isotretinoin Question, Even Though It Looks Like It Does

A study published in the October 2026 issue of the Journal of Investigative Dermatology reports that roughly a quarter of the genetic variants linked to acne are also linked to inflammatory bowel disease, and that the risk alleles point in the same direction for both conditions. The team, led by Willemijn C.A.M. Witkam and Sander Lamballais with senior author Kelsey R. van Straalen at Erasmus MC in Rotterdam, drew on genome-wide association data from more than 34,000 acne cases and 30,000 IBD cases to map 36 shared genomic regions, most of which had never been tied to acne before.

What the paper actually measures matters here. Cross-trait GWAS compares inherited DNA variants across two conditions and checks whether the same alleles raise or lower risk for both. That is a statement about shared germline biology, not about what happens when a person takes a drug. Concordant effect direction, the pattern the authors report, is the signature researchers use to argue for a common causal pathway rather than coincidence or confounding, but it does not by itself prove that one disease causes the other, or that a treatment for one disease triggers the other.

That distinction lands squarely on a decades-old controversy. Isotretinoin, the acne drug sold as Accutane, has been suspected of triggering inflammatory bowel disease since case reports emerged shortly after its 1982 FDA approval. A 2010 case-control study found a roughly 68% increased risk of ulcerative colitis among isotretinoin users, with Roche pulling the brand-name product in 2009 amid litigation tied partly to that concern. Because germline genetic variants are fixed at birth and cannot be caused by a prescription taken decades later, this new paper cannot be read as evidence about isotretinoin's safety one way or the other. It speaks to why acne and IBD co-occur, not to why some acne patients on a particular drug develop bowel symptoms.

Larger, more recent pharmacoepidemiology has already been chipping away at the isotretinoin theory on its own terms. One large-scale comparison found that Crohn's and ulcerative colitis risk was similar between isotretinoin users and patients on oral antibiotics for acne, suggesting the drug itself was not the distinguishing factor, though UC risk was transiently elevated in the first six months of treatment, a small absolute effect of roughly five extra cases per 10,000 patients. A propensity-matched cohort study using TriNetX data went further, finding that acne itself, not tetracycline antibiotics and not isotretinoin, carried the elevated odds of incident IBD (OR 1.42, CI 1.23 to 1.65). The new genetic data lines up with that drug-agnostic conclusion using a method immune to the confound that dogged the clinical literature for forty years.

The paper also has a direct scientific predecessor worth naming. An April 2022 Mendelian randomization study had already suggested that genetic liability to acne increases IBD risk, using statistical instruments to approximate a controlled experiment from observational data. The Rotterdam group's cross-trait analysis is best understood as a methodological upgrade to that earlier signal, swapping instrument-based inference for direct locus-level mapping across 36 shared regions, rather than a first discovery that acne and IBD share biology.

The pathway evidence the study surfaces, centered on T-cell activity and cytokine signaling, sits inside a pattern dermatology genetics has been building for years. Hidradenitis suppurativa, a severe follicular disease related to acne, already has an established genetic and clinical correlation with Crohn's disease, and psoriasis remains the reference case for skin-gut immune overlap through shared loci such as IL23R. Dermatology Times covered the study's clinical implications, framing severe or treatment-resistant acne as a potential signal for gastrointestinal screening along the lines already used for psoriatic arthritis, though that extrapolation goes beyond what a pleiotropy map can establish on its own.

There is a limitation the paper does not resolve. Population data from Taiwan's National Health Insurance Research Database has shown acne patients carry elevated rates of peptic ulcer disease, irritable bowel syndrome, GERD, and constipation, a broader gut-skin signature that extends well past IBD specifically. A cross-trait GWAS built around two conditions cannot say whether the shared loci it finds are specific to the acne-IBD pairing or are markers of a wider inflammatory tendency that happens to show up across several gut diagnoses at once. Until someone runs the same analysis against that fuller list of comorbidities, the 36 shared regions describe a correlation between two labels, not a bounded biological mechanism.