A University of Michigan team has split two lookalike drug rashes into separate immune diseases by mapping 59,608 skin cells across 18 biopsies. The verdict lands online September 10, 2026, in the Journal of Investigative Dermatology.

Picture an older male patient who took a beta-lactam antibiotic last week. He walks into clinic with sharp red patches in the groin and under the arms. Looks like a textbook drug rash. But is it morbilliform drug eruption (MDE), or the rarer symmetrical drug-related intertriginous and flexural exanthema (SDRIFE)?

The two can be told apart clinically, sometimes. Yet the molecular question — what is the immune engine underneath — has lingered for years. The Michigan group set out to answer it cell by cell.

How big was the dataset?

The team analyzed 59,608 cells drawn from 18 biopsies. That breaks down to 9,778 cells from healthy donors, 37,764 from MDE patients, and 12,066 from SDRIFE patients. MDE biopsies ran roughly three for every SDRIFE sample in the cohort.

They paired single-cell RNA sequencing with exploratory spatial transcriptomics and immunohistochemistry. SDRIFE cases were diagnosed using the 2004 Häusermann clinical criteria. Both eruptions were most often triggered by beta-lactam antibiotics.

What did the immune map actually show?

For MDE, the dominant cell type was CD8+ cytotoxic T cells, joined by Th1-type CD4+ T cells. The team found high expression of CXCL9, CXCL10, and CXCL11 — chemokines that pull more immune cells in. Interferon-responsive genes lit up across the lesion.

SDRIFE presented a different picture. CD8+ infiltration stayed limited. CD4+ T cells dominated, with only modest Th2 skewing. Macrophages carried an immunomodulatory signature. Spatial transcriptomics confirmed substantial T cell and macrophage presence in the superficial dermis for both eruptions — but the cellular flavor ran in opposite directions.

The authors concluded that MDE was characterized by cytotoxic and interferon-mediated inflammation, while SDRIFE involved an immunomodulatory, non-classical response. They pushed back on the older assumption, noting that "both conditions have been previously hypothesized to share similar immunopathogenic mechanisms."

Bottom line: the same drug class can drive two rashes on two different tracks.

Should the "drug rash" playbook change?

Different immune programs invite different therapeutic strategies. A cytotoxic, interferon-rich MDE lesion might respond to drugs that dampen CD8+ activity or block IFN signaling. A SDRIFE lesion, anchored by immunomodulatory macrophages and CD4+ T cells, may need a gentler approach — or none at all, since SDRIFE typically clears once the offending drug is stopped.

Laura B. Chopp, Xianying Xing, Rachael Bogle, Jennifer Fox, Joseph Kirma, J. Michelle Kahlenberg, Lam C. Tsoi, May P. Chan, and Johann E. Gudjonsson led the work out of Michigan. The takeaway for now: a one-size-fits-all playbook for "drug rash" no longer fits the biology.

What comes next?

The dataset is now live in the NCBI Gene Expression Omnibus under accession GSE309154. Any lab can pull the cells and interrogate them. Funding came from the A. Alfred Taubman Medical Research Institute and NIH grants P30-A075043 and P01-AI179251 (DOI: 10.1016/j.jid.2026.08.011).

The authors closed the paper by noting that the data highlight potential diagnostic and therapeutic targets for SDRIFE. Translation of single-cell snapshots into clinically actionable diagnostic panels now sits firmly on the to-do list.