Clinical evaluation of type 2 disease status in a real-world population of difficult to manage asthma using historic electronic healthcare records of blood eosinophil counts

Abstract

Background: Blood eosinophil measurement is essential for the phenotypic characterization of patients with difficult asthma and in determining eligibility for anti-IL-5/IL-5Rα biological therapies. However, assessing such measures over limited time spans may not reveal the true underlying eosinophilic phenotype, as treatment, including daily oral corticosteroid therapy, suppresses eosinophilic inflammation and asthma is intrinsically variable. Methods: We interrogated the electronic healthcare records of patients in the Wessex AsThma CoHort of difficult asthma (WATCH) study (UK). In 501 patients being evaluated in this tertiary care centre for difficult to control asthma, all requested full blood count test results in a 10-year retrospective period from the index WATCH assessment were investigated (n = 11,176). Results: In 235 biological therapy-naïve participants who had 10 or more measures in this time period, 40.3% were eosinophilic (blood eosinophils ≥300 cells/µl) at WATCH enrolment whilst an additional 43.1%, though not eosinophilic at enrolment, demonstrated eosinophilia at least once in the preceding decade. Persistent eosinophilia was associated with worse post-bronchodilator airway obstruction and higher Fractional exhaled Nitric Oxide (FeNO). In contrast, the 16.6% of patients who never demonstrated eosinophilia at this blood eosinophil threshold showed preserved lung function and lower markers of Type 2 inflammation. Conclusions: This highlights the central role that type 2 inflammation, as indicated by blood eosinophilia, has in difficult asthma and suggests that longitudinal electronic healthcare record analysis can be an important tool in clinical asthma phenotyping, providing insight that may help understand disease progression and better guide more specific treatment approaches.