Asthma is a heterogeneous disease, and most of its clinical phenotypes cannot identify distinct entities. Biomarker analysis may help in tailoring treatment and predicting the future risk of exacerbation in patients with severe asthma5,28,29. In this prospective observational study, we demonstrated that previous history of severe-to-serious exacerbation, baseline serum tryptase and TSLP levels, and blood eosinophil counts could independently predict the future development of exacerbation in patients with severe asthma. Most importantly, patients with severe asthma with high blood eosinophil counts and low serum tryptase levels were more likely to have greater risk of exacerbation than those with low blood eosinophil counts and high serum tryptase levels despite treatment with ICS-contained multiple therapy. The present study was the first to analyze the biomarkers of severe asthma comprehensively and proposed the novel idea that the possible combination of elevated TSLP levels and reduced tryptase levels might result in ongoing eosinophilia and non-responsiveness to high-dose ICS treatment.
High blood eosinophil count is associated with disease severity and eosinophilic airway inflammation in asthma30. In the Copenhagen General Population Study, Signe el al.31 reported that increased incidence of moderate-to-severe exacerbation is more strongly associated with high blood eosinophil counts (>290 cells/mL) than with low blood eosinophil counts (<180 cells/uL). In the UK, Price et al.24 reported that patients with asthma with blood eosinophil counts >400 cells/μL experience more severe exacerbations and have poorer asthma control. Our findings were consistent with these results. Absolute blood eosinophil counts of ³291 cells/mL had a higher probability of asthma exacerbation (AOR = 6.04). In addition, the patients with high blood eosinophil counts and low serum tryptase levels (≤1448 pg/mL) simultaneously suffered from more frequent exacerbations than those with low blood eosinophils and high tryptase levels (AOR up to 16.92).
Mast cells are tissue-based inflammatory cells of hematopoietic origin that respond to signals of innate and adaptive immunity. Mast cells play an important role in allergic diseases, including anaphylaxis, allergic rhinitis, and allergic asthma32. Human mast cells secrete α- and β-tryptases. Mature β-tryptase, which was measured in this study, is the predominant form stored in the secretory granules of mast cells. Tryptase is a specific marker of mast-cell activation, and thus tryptase levels can be reasonably measured to reflect the burden of mast cell activation in the allergic TH2 pathway in asthma19,32. Gao et al.20 reported that serum baseline tryptase levels in childhood asthma, as well as asthma control, serum IgE and IL-13 levels, blood eosinophil counts, and lung function parameters, are strongly correlated with disease severity of asthma. The Severe Asthma Research Program also reported that severe asthma is associated with the predominance of tryptase+chymase+mast cells in the airway submucosa and epithelium33. In addition, the gene expression of mast cell tryptase is increased in asthmatic epithelium, especially in the TH2-high subgroup, and predicts the responsiveness to ICS34. The numbers of airway tissue mast cells and the concentration of bronchoalveolar lavage tryptase can determine the efficacy of ICS treatment in persistent asthma35. The findings of our study were consistent with those of a previous study, which suggested that tryptase may be implicated in steroid responsiveness in the asthma inflammatory process and may determine the future risk of asthma exacerbation.
TSLP, which is produced mainly by the lung and gut epithelia, skin keratinocytes, and dendritic cells, is involved in various allergic diseases, including bronchial asthma, atopic dermatitis, and eosinophilic esophagitis. TSLP expression can be induced by several cytokines, respiratory viruses, bacterial and fungal products, allergens, cigarette smoke extracts, and tryptase36. In asthma, increased TSLP concentrations are observed in bronchoalveolar lavage, induced sputum, exhaled breath condensate, and plasma37–40. TSLP expression is increased in the airway mucosa in a subset of severe asthmatics despite high-dose inhaled or oral steroid treatment41. TSLP can induce steroid resistance and abrogate the inhibitory effects of dexamethasone on type 2 cytokine production in ILC2 cells42. In the present study, we found that TSLP per se is an independent factor for predicting future risk of asthma exacerbation, and serum TSLP levels ≥25 pg/mL are associated with a high probability of asthma exacerbation (AOR = 8.19). Unsurprisingly, Corren et al.43 reported that anti-TSLP monoclonal antibody reduces annual exacerbation rates by 62%–71% at different doses in uncontrolled asthma despite treatment with long-acting b2 agonists and medium-to-high doses of ICS. Their findings have suggested some biological plausibility for TSLP being a contributor and an indicator of asthma exacerbation. Collectively, serum TSLP may contribute to steroid resistance, whereas tryptase may suggest steroid responsiveness in asthma inflammatory process, as observed in the present study. Moreover, our study suggested the novel idea that the possible combination of elevated TSLP levels and reduced tryptase levels might result in ongoing eosinophilia and non-responsiveness to high-dose ICS treatment. This combination of biomarkers (high TSLP levels and low tryptase levels) might indicate that these patients with severe asthma are suitable for anti-TSLP therapy.
The previous history of severe-to-serious exacerbation is an independent factor predicting future exacerbation (AOR = 3.27). This result was consistent with that of a previous study that recent severe asthma exacerbations are an important independent predictor of future severe exacerbation in children with severe/difficult-to-treat asthma44. Similarly, a prospective analysis of patients aged ≥12 years with severe/difficult-to-treat asthma indicated that recent severe asthma exacerbations appear to be a strong independent factor predicting future exacerbations (AOR = 3.77)45. These findings should prompt physicians to understand the contributing factors and pathological process driving these exacerbations and refine asthma management to prevent future exacerbation.
Our study has several limitations. First, serial examination of serum biomarkers was not performed to delineate the relationship between changes in biomarkers and asthma control status. Second, all study subjects were under maintenance treatment. Multi-treatment might have influenced the levels of the biomarkers at the initiation of the study. Third, this study was observational in nature, and replicating the results in another cohort is needed. Furthermore, whether the strategy to reduce serum TSLP levels, serum tryptase levels, or blood eosinophil counts in these patients with severe asthma can reduce future development of asthma exacerbation remains to be tested. Therefore, further validation must be performed. Nevertheless, the estimated power (1-β) was 0.99 for our sample size.