Integrative epidemiology and immunotranscriptomics uncover a risk and potential mechanism for cutaneous lymphoma unmasking or progression with dupilumab therapy.

backgroundThere have been multiple reports of the anti-IL-4Rα agent, dupilumab, being associated with the onset and/or progression of cutaneous T-cell lymphoma (CTCL).

objectiveWe sought to evaluate safety signals associated with dupilumab, with a focus on CTCL, and to evaluate the possible underlying mechanism or mechanisms for the potential association.

methodsFirst, we used the Food and Drug Administration's pharmacovigilance database, FAERS (FDA Adverse Event Reporting System), to evaluate whether dupilumab was associated with CTCL, including both positive outcome controls (conjunctivitis, eosinophilia, and arthralgia) and exposure controls (other medications with similar indications, including JAK inhibitors and the anti-IL-13 agent, tralokinumab) to evaluate confounding bias. Thereafter, we used publicly available bulk and single-cell RNA sequencing datasets to probe possible underlying mechanisms through which dupilumab might be associated with CTCL.

resultsBetween January 2017 and the fourth quarter of 2023, there were 181,575 unique reports of dupilumab-related adverse events (AEs) in FAERS, with 606 of these being for a neoplasm. Dupilumab had 30.0 times the proportional reporting ratio (PRR) (95% confidence interval, 25.0-35.9) for CTCL compared to all other medications in FAERS. The risk was highest in men aged 45 to 65. The PRR for conjunctivitis, eosinophilia, and arthralgia, known adverse effects of dupilumab, were 35.6 (34.4-36.8), 2.15 (2.00-2.31), and 2.14 (2.07-2.18), respectively. Using the log-count normalized PRR (AE score) to account for PRR inflation when reports were small, the top safety signals included conjunctivitis (AE score 8.3) and CTCL (AE score 4.9). Bulk RNA sequencing data showed changes in IL-4RA and IL-13RA1 expression in CTCL and in epidermal layers of atopic dermatitis (AD) biopsy samples. Single-cell transcriptomic studies revealed that this change was similar in AD and CTCL, and that keratinocytes seemed to be the most divergent cell type with regards to IL-4R and IL-13RA1. An effect on keratinocyte-specific gene expression was also independently observed in available bulk RNA sequencing data.

conclusionThese data suggest that dupilumab might be causing an unmasking or progression of CTCL via the same mechanism through which it improves AD: IL-13 receptor blockade, which leads to increased IL-13 in the local milieu, driving CTCL stimulation and progression. However, these associations need further evaluation given the inherent limitations of the FAERS database and our nonexperimental approach.