Background and objectives: Janus kinase (JAK) inhibitors are emerging as a therapeutic option for alopecia areata. The risk of potential adverse events is currently debated. In particular, several safety data for JAK inhibitors are extrapolated from a single study in elderly patients with rheumatoid arthritis treated with tofacitinib or adalimumab/etanercept as a comparator. The population of patients with alopecia areata is clinically and immunologically different from persons with rheumatoid arthritis and tumor necrosis factor (TNF) inhibitors are not effective in these patients. The objective of this systematic review was to analyze available data on the safety of various JAK inhibitors in patients with alopecia areata.

Methods: The systematic review was performed according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A literature review was performed by searching PubMed, Scopus and EBSCO databases with the last search on March 13, 2023.

Results: In total, 36 studies were included. The frequency and odds ratio (OR) for most common adverse events versus placebo were: for baricitinib hypercholesterolemia (18.2% vs 10.5%, OR = 1.9) and headache (6.1% vs 5.1%, OR = 1.2), for brepocitinib elevated creatinine level (27.7% vs 4.3%, OR = 8.6) and acne (10.6% vs 4.3%, OR = 2.7), for ritlecitinib acne (10.4% vs 4.3%, OR = 2.6) and headache (12.5% vs 10.6%, OR = 1.2) and for deuruxolitinib headache (21.4% vs 9.1%, OR = 2.7) and acne (13.6% vs 4.5%, OR = 3.3). The respective numbers for upper respiratory infections were: baricitinib (7.3% vs 7.0%, OR = 1.0) and brepocitinib (23.4% vs 10.6%, OR = 2.6); for nasopharyngitis: ritlecitinib (12.5% vs 12.8%, OR = 1.0) and deuruxolitinib (14.6% vs 2.3%, OR = 7.3).

Conclusions: The most common side effects of JAK inhibitors in patients with alopecia areata were headache and acne. The OR for upper respiratory tract infections varied from over 7-fold increased to comparable to placebo. The risk of serious adverse events was not increased.