Serum metabolomic profiling identifies taurolithocholic acid as a predictor of HDM-SCIT response in allergic rhinitis: clinical discovery and experimental validation.

Background: Subcutaneous immunotherapy (SCIT) is a clinically effective and disease-modifying treatment for house dust mite (HDM)-induced allergic rhinitis (AR). Although SCIT provides long-term symptom improvement for most patients, a subset of patients shows inadequate clinical response. The underlying factors contributing to the heterogeneity in treatment efficacy remain unclear, and robust biomarkers capable of predicting SCIT outcomes are still lacking. Objective: This study aimed to evaluate the clinical efficacy of one-year HDM-SCIT and to identify serum biomarkers associated with therapeutic response, with the goal of improving prediction of SCIT outcomes in patients with AR. Methods: Eligible patients with HDM-induced allergic rhinitis were prospectively enrolled from two medical centers (Peking Union Medical College Hospital and the Affiliated Hospital of Qingdao University) and subsequently completed one year of HDM-SCIT. Symptom severity was evaluated using the visual analog scale (VAS) and total nasal symptom score (TNSS), and quality of life by the rhinoconjunctivitis quality of life questionnaire (RQLQ) at baseline and one year after treatment. Clinical remission was defined as ≥ 30% improvement of VAS and RQLQ scores. In addition, medication scores and TNSS were also included as secondary endpoints to support the response definition. Paired serum samples collected before and after treatment were analyzed using untargeted metabolomics and targeted bile acid profiling based on liquid chromatography-mass spectrometry. To validate metabolomic findings, an HDM-induced AR mouse model was established, followed by SCIT alone or SCIT combined with taurolithocholic acid (TLCA, a bile acid). Nasal histopathology, T cell subsets in the spleen, and cytokine levels in serum and nasal lavage fluid were assessed. Results: Of the 32 patients who received one-year HDM-SCIT, 22 were responders and 10 were non-responders. No baseline differences in age, AR duration, comorbidities, total IgE, VAS, TNSS, or RQLQ were found between the two groups. However, responders had significantly greater reductions in VAS, TNSS, and RQLQ post-one-year treatment (p < 0.05). Untargeted metabolomics identified 956 differential metabolites (614 upregulated, 342 downregulated) between the pre-treatment group and post-treatment group, 1,389 (582 up, 807 down) and 1574 (342 up, 1232 down) between responders and non-responders at baseline and post-treatment, respectively. KEGG analysis highlighted bile secretion as a key differential pathway. Bile acid targeted metabolomics revealed TLCA as a potential biomarker associated with HDM-SCIT efficacy. Mouse models confirmed that HDM-SCIT combined with TLCA, particularly at high dose, alleviated nasal mucosal inflammation (reduced epithelial damage, eosinophils), increased splenic CD4 Conclusion: Serum TLCA levels were associated with clinical response to HDM-SCIT. Animal model validation demonstrated that TLCA may enhance SCIT-induced immune tolerance. These findings support TLCA may serve as a potential metabolite-based biomarker and adjunct target for improving the efficacy of allergen immunotherapy.