通过非经典2型炎症轴识别过敏性鼻炎中的免疫负荷状态及预测因素

文献信息

中文摘要

背景： 虽然经典2型（T2）炎症因子在过敏性鼻炎（AR）中的作用已被充分表征，但非经典T2炎症因子的作用研究较少。本研究旨在阐明非经典T2炎症因子在AR中的作用及预测意义。
方法： 从GEO检索AR相关数据集进行生物信息学分析。共纳入64例AR患者和42例健康对照。采用ELISA检测非经典T2因子和T1细胞因子。使用SNOT-22和VAS评分评估AR疾病严重程度。分别进行PCA-PAM聚类、MCA和相关性网络分析，以探讨非经典T2因子的炎症模式和相互作用。构建ROC曲线以评估预测性能。在PCA、MCA和ROC分析之前，所有生物标志物均以健康对照中位数和IQR进行稳健标准化。通过LOOCV、bootstrap重采样和10折交叉验证进行内部验证。
结果： AR患者血清ALOX15以及ALOX15、CLC、CST1、POSTN、TSLP和IL-33与T1细胞因子的比值显著升高，且ALOX15在鼻黏膜和血清中转录本和蛋白水平均一致上调。基于PCA衍生的T1/T2生物标志物的PAM聚类将AR患者分为低免疫应答型（LIR-AR，n=40）和高免疫应答型（HIR-AR，n=24）内型。MCA确定CST1、POSTN和CLC为最主要的判别因子，IL-33在空间上与下游效应物分离。CST1水平和ALOX15/IFN-γ比值分别与流涕和鼻痒的VAS评分相关。ALOX15/TNF-α比值对AR表现出良好的预测准确性（AUC=0.856±0.106）。
结论： AR以T1/T2失衡为特征，鼻黏膜和血清中ALOX15升高且T2/T1比值增加，提示ALOX15可能是一种潜在生物标志物。CST1、POSTN和CLC是区分LIR-AR与HIR-AR的核心变量，可能有助于免疫分型鉴别。CST1表达和ALOX15/IFN-γ比值分别与流涕和鼻痒相关；然而，这些基于相关性的发现就因果关系而言需谨慎解读。ALOX15/TNF-α比值对AR显示出良好的诊断价值，这些探索性发现需要在更大规模独立队列和功能研究中验证。

英文摘要

BACKGROUND: While the roles of classical type 2 (T2) inflammatory factors in allergic rhinitis (AR) have been well characterized, the roles of non-classical T2 inflammation factors are less explored. This study aimed to elucidate the roles and predictive significance of non-classical T2 inflammatory factors in AR.
METHODS: AR-related datasets were retrieved from GEO for bioinformatics analysis. A total of 64 AR patients and 42 healthy controls were included. Non-classical T2 factors and T1 cytokines were measured by ELISAs. The SNOT-22 and VAS scores were used to evaluate the disease severity of AR. PCA-PAM clustering, MCA and correlation network analysis were performed to explore the inflammatory pattern and interactions of non-classical T2 factors, respectively. ROC curves were constructed to evaluate the predictive performance. All biomarkers were robustly standardized to healthy control medians and IQRs prior to PCA, MCA, and ROC analyses. Internal validation was performed via LOOCV, bootstrap resampling, and 10-fold cross-validation.
RESULTS: Serum ALOX15 and ratios of ALOX15, CLC, CST1, POSTN, TSLP, and IL-33 to T1 cytokines were significantly elevated in AR patients, with concordant upregulation of ALOX15 at transcript and protein levels in both nasal mucosa and serum. PAM clustering of PCA-derived T1/T2 biomarkers stratified AR patients into the low-immune-response type (LIR-AR, n = 40) and the high-immune-response type (HIR-AR, n = 24) endotypes. MCA identified CST1, POSTN, and CLC as the top discriminators, with IL-33 spatially separated from downstream effectors. CST1 levels and the ALOX15/IFN-γ ratios were correlated with the VAS scores for rhinorrhea and nasal itching, respectively. The ALOX15/TNF-α ratio exhibited good predictive accuracy for AR (AUC = 0.856 ± 0.106).
CONCLUSIONS: AR was characterized by T1/T2 imbalance, with elevated ALOX15 and increased T2/T1 ratios in both nasal mucosa and serum, suggesting ALOX15 as a potential biomarker. CST1, POSTN, and CLC were core variables distinguishing LIR-AR from HIR-AR and may be useful for immunotype differentiation. CST1 expression and the ALOX15/IFN-γ ratio correlated with rhinorrhea and nasal itching, respectively; however, these correlation-based findings require cautious interpretation regarding causality. The ALOX15/TNF-α ratio showed good diagnostic value for AR, with these exploratory findings requiring validation in larger independent cohorts and functional studies.