Serum biomarkers including periostin and eosinophils for predicting clinical remission in severe asthma: A systematic review

Background: Clinical remission is an emerging, ambitious treatment goal in severe asthma, characterized by sustained absence of symptoms and exacerbations, and normalization of lung function while off or on minimal treatment. Easily accessible predictors are needed to guide therapy. Serum biomarkers represent promising candidates. Objective: To systematically review and synthesize the evidence on the association between serum biomarker levels and the achievement of clinical remission in patients with severe asthma. Methods: We searched PubMed, Embase, and the Cochrane Central Register of Controlled Trials from inception to October 2025 for observational studies and clinical trials reporting serum biomarkers and a predefined clinical remission outcome in severe asthma. Study quality was assessed using the Newcastle-Ottawa Scale. Results: 18 studies (n=4,250 patients) were included. Key biomarkers studied were serum eosinophil count (sEOS), periostin, IgE, and a limited number of emerging markers, including soluble IL-5 receptor α (sIL-5Rα). Higher baseline sEOS was consistently associated with a greater likelihood of achieving remission on biologic therapy; across studies, reported ORs for sEOS ranged from approximately 1.5 to 2.9, with a narrative summary estimate of about 2.1. Serum periostin showed a modest positive association, while total IgE was not consistently predictive. Evidence for IL-5 and IL-13 cytokines themselves was scarce, whereas soluble IL-5 receptor α (sIL-5Rα) showed the most promising emerging signal, although based on only a few studies. Conclusion: Current evidence supports serum eosinophil count as the most consistent predictor of clinical remission in severe asthma, with periostin as a secondary biomarker. Evidence for IL-5 and IL-13 remains limited, while sIL-5Ra appears promising but requires further validation. Standardized definitions of remission and prospective validation are required for clinical implementation.