Feno differentiates epithelial gene expression clusters: Exploratory analysis from the MESOS randomized controlled trial

Background Understanding how asthma biomarkers relate to gene expression signatures could help identify drivers of pathogenesis. Objective This post hoc exploratory analysis of the phase II tralokinumab trial MESOS (ClinicalTrials.gov identifier NCT02449473) aimed to profile baseline airway inflammation in patients with moderate-to-severe asthma. Methods The T2 and T17 gene expression signatures, 3-gene mean and 5-gene mean, were calculated through transcriptomic analysis of baseline bronchial brushing samples. Clustering analysis using these signatures identified 3 distinct inflammatory subgroups: T2LOW/T17HIGH (n = 33), T2HIGH/T17LOW (n = 10), and T2LOW/T17LOW (n = 27). Results Fractional exhaled nitric oxide (Feno) levels were highest for T2HIGH/T17LOW and lowest for T2LOW/T17HIGH (median = 52.0 [range 42.5-116.3] and median = 18.8 [range 6.6-128.6] ppb, respectively; P = .003). High Feno levels were strongly correlated with high T2 gene expression (Spearman ρ = 0.5537; P < .001). Individual genes differentially expressed in patients with elevated levels of Feno, blood and bronchial submucosal eosinophil counts, and IgE level were explored, with cystatin SN (CST1) being the most upregulated gene in all subgroups (4.49- to 34.42-fold upregulation across clinically defined subgroups with high biomarker expression). Conclusion Feno level may be useful to differentiate patients with T2 or T17 gene expression. Elevated Feno levels are associated with high CST1 expression.