Portal de Periódicos da CAPES

Type 2 biomarker expression (FeNO and blood eosinophils) is higher in severe adult‐onset than in severe early‐onset asthma

2021; Wiley; Volume: 76; Issue: 10 Linguagem: Inglês

10.1111/all.14997

1398-9995

Marek Lommatzsch, Maria Klein, Paul Stoll, J. Christian Virchow,

IL-33, ST2, and ILC Pathways

Severe asthma is predominantly associated with a type 2 inflammatory pattern1; however, it is unclear whether the expression of type 2 biomarkers differs between severe early-onset asthma (SEA) and severe adult-onset asthma (SAA). This issue is of particular importance for the question whether the age of asthma onset can be a helpful clinical information for the selection of targeted treatments (especially biologics) in patients with severe asthma. We analyzed data from patients with severe asthma (not currently treated with biologics) who presented routinely to a specialized severe asthma outpatient clinic in Rostock (Germany). Inclusion criteria and methods are detailed in the Supplement. According to GINA, presence of type 2 inflammation was defined as ≥150 eosinophils/µl blood and/or FeNO ≥20 ppb and/or a history of clinically allergen-driven symptoms.2 According to the UK Severe Asthma Registry (UKSAR),1 a type 2 high phenotype was defined as ≥150 eosinophils/µl blood and FeNO ≥25 ppb, a type 2 low phenotype as <150 eosinophils/µl blood and FeNO <25 ppb. A type 2 intermediate phenotype was defined by the detection of either ≥150 eosinophils/µl blood or FeNO ≥25 ppb. A total of 220 adult patients were included: 50 with SEA (symptom onset <18 years of age) and 170 with SAA (symptom onset ≥18 years of age). There were no differences in asthma control, gender distribution, body mass index, history of upper airway diseases, total IgE, inhalation therapies, lung function and diffusion capacity between the groups (Tables S1 and S2). As compared to SEA, SAA was characterized by a higher smoking history, more exacerbations and a higher likelihood of maintenance OCS treatment (Table S1). Patients with SAA had significantly higher FeNO values and blood eosinophils than patients with SEA (Table S2, Figure 1). The correlation between blood eosinophils and FeNO was weak (r = .18 in both groups; SEA: p = .24, SAA: p = .02). A type 2 high phenotype (UKSAR definition1) was more prevalent in SAA than in SEA (Table S2 and Figure 2). In both groups, <10% of the patients had a type 2 low phenotype according to UKSAR1 (Figure 2), and 40 years at disease onset (later adulthood). Although there was a trend to higher FeNO values and blood eosinophils in patients with a disease onset in later adulthood, there were no significant differences in type 2 biomarkers between these SAA subgroups (Table S3). This is the first report showing that FeNO values are higher in SAA than in SEA (previous studies investigated FeNO in various patient clusters, but did not compare SEA with SAA3). In addition, although it is generally accepted that intrinsic asthma is associated with higher blood eosinophils than allergic asthma4 and that SAA is characterized by stronger tissue and sputum eosinophilia than SEA,3, 5 our study demonstrates stronger blood eosinophilia in SAA than in SEA. Accordingly, we show that patients with SAA are more likely to meet current criteria of a type 2 high phenotype1 than patients with SEA. Thus, we hypothesize that SAA is characterized by an intrinsically stronger type 2 inflammation than SEA, reflecting the clinical observation that adult-onset asthma tends to be more severe than early-onset asthma. In addition, the data suggest that the age of asthma onset might serve as a helpful clinical information for the selection of targeted treatments. Indeed, the adult-onset characteristic is already used by clinicians in choosing biologic therapies targeting the Interleukin-5 (IL-5) pathway, because of post hoc data from clinical trials,6 real-world evidence studies7 and current GINA recommendations on the use of biologics.2 Our study revealed a remarkable association of a higher type 2 biomarker expression (FeNO, blood eosinophils) with a lower probability of allergies (and at least a trend to lower total IgE), both in the comparison between SAA and SEA, and in SAA subgroup analyses. This finding is supported by recent data showing that eosinophilia in patients with SAA is largely independent from IgE-mediated pathways.8 However, allergic inflammation can promote NO production and eosinophilia in allergic airway diseases.9, 10 Therefore, we hypothesize that the intrinsic type 2 inflammation in SAA is stronger than that driven by allergic inflammation in SEA. It needs to be acknowledged that the patients in our cohort had allergies to typical aeroallergens: it cannot be excluded that patients with less common allergies might have a higher type 2 biomarker expression. In addition, one has to bear in mind that blood eosinophils and FeNO values can vary over time and can be substantially influenced by changing doses of ICS.11 It is interesting to note that only a small minority (<10%) of our patients with severe asthma displayed a type 2 low phenotype: this in accordance with data from the UK1 and Denmark.12 Therefore, given the ongoing discussion whether ‘type 2 low asthma’ or ‘neutrophilic asthma’ exists as a genuine disease entity,13 more research is needed to better characterize patients with typical asthma symptoms and low type 2 markers. We compared type 2 biomarker expression between severe early-onset asthma (SEA: age of onset <18 years) and severe adult-onset asthma (SAA: age of onset ≥18 years) in 220 patients (not treated with biologics) routinely referred to a tertiary asthma centre (Rostock, Germany). We demonstrate that SAA is characterized by higher FeNO values (median: 41 vs. 20 ppb, p < .001) and higher blood eosinophils (median: 500 vs. 350 eosinophils/µl, p = .01) than SEA. Accordingly, a type 2 high phenotype (≥150 eosinophils/µl blood and FeNO≥25 ppb) is more prevalent in SAA than in SEA (64.7% vs. 38%, p = .002). Thus, we conclude that SAA is characterized by a stronger type 2 phenotype than SEA, suggesting that the age of onset might be a helpful clinical information for the identification of targeted treatments. None. The authors report no conflicts of interest regarding this report. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.