Janus kinase inhibitors in atopic dermatitis: an umbrella review of meta-analyses

Background: Clinicians and healthcare policymakers have been drenched with a deluge of overlapping meta-analyses (MAs), and the necessity for comprehensive and clearly defined evidence of Janus kinase inhibitors (JKIs) in atopic dermatitis (AD) is urgent. Methods: Six databases were searched for MAs published until October 2023. Qualitative description of MAs was mainly used, and Investigator's Global Assessment response (IGA response), the 75% improvement in Eczema Area and Severity Index (the EASI75), peak pruritus Numerical rating score (PP-NRS), and adverse effects were cited to describe the efficacy and safety of JKIs. The methodological quality of the included MAs was assessed by A Measurement Tool to Assess Systematic Reviews II (AMSTAR II), and the quality of evidence was evaluated by the grading of recommendations, assessment, development, and evaluation (GRADE). Results: Sixteen MAs were pooled in this review, of which five studies appraised JKIs, five appraised systemic JKIs, five papers assessed abrocitinib only, and one assessed baricitinib. Two studies were of "high" methodological quality and 14 MAs were of "moderate" quality. Eleven MAs integrated the results of JKIs and reported that JKIs provide faster onset of IGA response (RR=2.83, 95% CI [2.25, 3.56], high-quality evidence). Similarly, 10 MAs showed that JAK inhibitors were more effective in improving the EASI75 (RR=2.84, 95% CI [2.2, 3.67], high-quality evidence). Results from 12 MAs showed JKIs were active in reducing the PP-NRS (SMD=-0.49, 95% CI [-0.67, -0.32]). All MAs affirmed JKIs added no adverse effects leading to discontinuation and serious adverse events (P<0.05). However, 200mg of abrocitinib had a higher risk of acne (RR=4.34, 95% CI [1.61, 11.71), herpes zoster (RR=1.64, 95% CI [0.42, 6.39]), headache (RR=1.76, 95% CI [1.03, 3]), and nausea (RR=7.81, 95% CI [3.84, 15.87]). Upadacitinib was known to increase acne (RR=6.23, 95% CI [4.08, 9.49]), nasopharyngitis (RR=1.36, 95% CI [1.03, 1.8]) and blood creatine phosphokinase (blood CPK) (RR=2.41, 95% CI [1.47, 3.95]). Baricitinib at 2mg was associated with increased blood CPK (RR=2.25, 95% CI [1.1, 2.97]). Conclusion: Compared to placebo or dupilumab, the administration of JKIs can ameliorate IGA response more effectively, improve the EASI75, and relieve pruritus without severe adverse effect, while accompanied by more acne, nasopharyngitis, headache, and digestive disturbances. The curative effect of 200 mg of abrocitinib is significant and more caution should be given in patients with gastrointestinal dysfunction, herpes zoster, and those who are acne-prone. Baricitinib and upadacitinib should be avoided in populations at high risk for cardiovascular events.