Department of Otolaryngology-Head and Neck Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Hubei Clinical Research Center for Nasal Inflammatory Diseases, Wuhan, China.
Allergy. 2024 May;79(5):1230-1241. doi: 10.1111/all.16068. Epub 2024 Feb 25.
Identifying predictive biomarkers for allergen immunotherapy response is crucial for enhancing clinical efficacy. This study aims to identify such biomarkers in patients with allergic rhinitis (AR) undergoing subcutaneous immunotherapy (SCIT) for house dust mite allergy.
The Tongji (discovery) cohort comprised 72 AR patients who completed 1-year SCIT follow-up. Circulating T and B cell subsets were characterized using multiplexed flow cytometry before SCIT. Serum immunoglobulin levels and combined symptom and medication score (CSMS) were assessed before and after 12-month SCIT. Responders, exhibiting ≥30% CSMS improvement, were identified. The random forest algorithm and logistic regression analysis were used to select biomarkers and establish predictive models for SCIT efficacy in the Tongji cohort, which was validated in another Wisco cohort with 43 AR patients.
Positive SCIT response correlated with higher baseline CSMS, allergen-specific IgE (sIgE)/total IgE (tIgE) ratio, and frequencies of Type 2 helper T cells, Type 2 follicular helper T (T2) cells, and CD23 nonswitched memory B (B) and switched memory B (B) cells, as well as lower follicular regulatory T (T) cell frequency and T/T2 cell ratio. The random forest algorithm identified sIgE/tIgE ratio, T/T2 cell ratio, and B frequency as the key biomarkers discriminating responders from nonresponders in the Tongji cohort. Logistic regression analysis confirmed the predictive value of a combination model, including sIgE/tIgE ratio, T/T2 cell ratio, and CD23 B frequency (AUC = 0.899 in Tongji; validated AUC = 0.893 in Wisco).
A T- and B-cell signature combination efficiently identified SCIT responders before treatment, enabling personalized approaches for AR patients.
鉴定变应原免疫治疗应答的预测性生物标志物对于提高临床疗效至关重要。本研究旨在鉴定尘螨过敏患者接受皮下免疫治疗（SCIT）后发生的此类生物标志物。
同济（发现）队列纳入 72 例完成 1 年 SCIT 随访的变应性鼻炎（AR）患者。在 SCIT 前采用多重流式细胞术对循环 T 和 B 细胞亚群进行特征分析。在 12 个月 SCIT 前后评估血清免疫球蛋白水平和联合症状和用药评分（CSMS）。确定 CSMS 改善≥30%的患者为应答者。采用随机森林算法和逻辑回归分析从同济队列中筛选出 SCIT 疗效的生物标志物并建立预测模型，然后在另一项纳入 43 例 AR 患者的 Wisco 队列中进行验证。
阳性 SCIT 应答与基线 CSMS、过敏原特异性 IgE（sIgE）/总 IgE（tIgE）比值较高，以及 2 型辅助 T 细胞、2 型滤泡辅助 T（T2）细胞和 CD23 未转换记忆 B（B）和转换记忆 B（B）细胞频率较高相关，与滤泡调节性 T（T）细胞频率和 T/T2 细胞比值较低相关。随机森林算法确定 sIgE/tIgE 比值、T/T2 细胞比值和 B 细胞频率是区分同济队列中应答者和非应答者的关键生物标志物。逻辑回归分析证实了包括 sIgE/tIgE 比值、T/T2 细胞比值和 CD23 B 细胞频率的组合模型的预测价值（同济队列的 AUC=0.899；Wisco 队列的验证 AUC=0.893）。
T 细胞和 B 细胞特征组合在治疗前有效地鉴定了 SCIT 应答者，为 AR 患者实现了个性化治疗方法。
