Answers are grounded in this paper and cite their sources.
Blood eosinophil counts have become the entry criterion for biologic therapy in chronic obstructive pulmonary disease (COPD). However, the eosinophil-selected programmes produced mixed results, and the 2026 alarmin trials reduced exacerbations without any eosinophil threshold. We re-read the published primary and subgroup estimates of 16 randomised trials and asked whether eosinophil gradients behave as predictive or merely prognostic markers. Within populations already enriched at 300 cells/µL, higher counts did not consistently predict larger effects: the rate ratio for counts of 500 cells/µL or more versus fewer was 0.51 versus 0.80 in BOREAS but 0.70 versus 0.65 in NOTUS. In contrast, exhaled nitric oxide of 20 ppb or more was associated with a larger dupilumab effect in the same direction in both trials (pooled approximate interaction p = 0.009). Tozorakimab reduced exacerbations across all eosinophil strata, with a gradient (0.77 to 0.57; approximate interaction p = 0.077) that is suggestive but statistically inconclusive. Astegolimab showed no modification by eosinophils or smoking status. We propose that blood eosinophils are a downstream messenger of an upstream epithelial–alarmin endotype rather than the therapeutic target itself. Blood eosinophils remain a simple and useful marker of risk and enrichment, but selection may need to move from a single-cell threshold to composite endotypes that combine exposure history, airway biomarkers and structural traits. Further studies are warranted, designed to test treatment-by-biomarker interactions prospectively.
No takes yet. Share an insight, caveat, or question.
Mari et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: