Biomarker algorithm may not accurately identify patients with eosinophilic asthma
Click Here to Manage Email Alerts
Key takeaways:
- A previously proposed algorithm uses noninvasive and accessible biomarkers to classify patients with severe asthma based on the likelihood of an eosinophilic phenotype.
- The researchers found no correlations between classifications based on the proposed biomarkers and those based on sputum analysis.
- Sputum analysis continues to be the gold standard for assessing inflammation of the airway.
A group of biomarkers proposed in previous research insufficiently diagnosed eosinophilic phenotypes among patients with asthma, according to a study published in Clinical and Translational Allergy.
Thus, sputum analysis remains the gold standard for assessing airway inflammation, Diana Betancor, MD, resident in the allergy department at Hospital Universitario Fundación Jiménez Díaz in Madrid, and colleagues wrote.
The algorithm — proposed by Heaney and colleagues in 2021 and published in Chest — used peripheral blood eosinophil (PBE) count, current treatment, fraction of exhaled nitric oxide, presence of nasal polyps and age of onset to classify patients with severe asthma based on the likelihood of an eosinophilic phenotype.
In the current retrospective observational study, Betancor and colleagues analyzed data from 145 patients with asthma aged 18 to 75 years (average age, 48 years) from eight Spanish hospitals who were participants in the MEGA cohort.
Based on the Heaney algorithm, 69.6% were grade 3 and 20.8% were grade 2, both considered likely eosinophilic; 16.8% were grade 1, or less likely to be eosinophilic; and 5.9% were grade 0, or non-eosinophilic. Patients classified as grade 3 had higher levels of FeNO, PBE and nasal polyps than the other groups (P < .005).
When the researchers compared these classifications with sputum eosinophilia results, however, they only found an agreement of 0.025 (95% CI, 0.013-0.037). The researchers noted a higher sputum eosinophilia rate among the grade 3 patients, but they added that it was not statistically significant compared with the other grades.
The researchers further noted that there were similar percentages of patients with sputum eosinophils greater than 3% in grades 1, 2 and 3, indicating that the proposed algorithm was unable to phenotype these patients.
Similarly, the researchers found no significant correlation between sputum eosinophils and PBE count in the full cohort (Spearman’s correlation coefficient = 0.11) or in the individual groups, with a Spearman’s correlation coefficient of –0.17 in grade 0, 0.08 in grade 1, 0.35 in grade 2 and 0.16 in grade 3.
Although the researchers said that there is an association between worse lung function and eosinophilic phenotypes, the current study did not find any differences in lung function, probably due to the presence of sputum eosinophilia in each grade.
Eosinophilic asthma has been correlated with greater severity and exacerbations as well, the researchers continued, although the higher exacerbation rates and severity that they found in the eosinophilic grades of this study did not reach statistical significance.
Heaney and colleagues chose accessible and noninvasive biomarkers for their algorithm, the researchers wrote, but PBE can be influenced by treatment and other illnesses, whereas age; gender; atopy; and tobacco, food and beverage consumption can affect FeNO.
Based on these findings, the researchers concluded that the biomarker groupings that Heaney and colleagues proposed are insufficient for diagnosing eosinophilic phenotypes in patients with asthma and that sputum analysis is still the “gold standard” for assessing airway inflammation.