Large-scale proteomic profiling identifies distinct inflammatory phenotypes in acute respiratory distress syndrome: a multicentre, prospective cohort study

作者：Mengna Lin, Feixiang Xu, Yiyu Deng, Ying Wei, Feng Shi, Yun Xie, Cuiying Xie, Chen X. Chen, Jianfeng Song, Yao Shen, Yiyan Lin, Hailin Ding, Yannan Zhou, Su Lü, Yumei Chen, Lulu Lan, Wenxin Zhao, Jing Zhu, Zhongshu Kuang, Wei Pang, Sijin Que, Xiaoyu Fang, Ran Ji, Chenyang Dong, Jiancheng Zhang, Qi Liu, Zhaocai Zhang, Chengjin Gao, Leilei Chen, Yuanlin Song, Liying Zhan, Lihong Huang, Xueling Wu, Ruilan Wang, Zhenju Song · 发表于：European Respiratory Journal · 年份：2025 · DOI：10.1183/13993003.00933-2025 · 被引用次数：9 · 研究领域：Respiratory Support and Mechanisms、S100 Proteins and Annexins、Immune Response and Inflammation

Background Host responses during acute respiratory distress syndrome are highly heterogeneous, contributing to inconsistent therapeutic outcomes. Proteome-based phenotyping may identify biologically and clinically distinct phenotypes to guide precision therapy. Methods In this multicentre cohort study, we used latent class analysis of targeted serum proteomics to identify acute respiratory distress syndrome phenotypes. Serum samples were collected within 72 h of diagnosis to capture early-phase profiles. Validation was conducted in external cohorts. Pathway enrichment assessed molecular heterogeneity. Lung computed tomography scans were analysed using machine learning-based radiomics to explore phenotypic distinctions. Heterogeneous treatment effects for glucocorticoids and ventilation strategies were evaluated using inverse probability of treatment weighting adjusted Cox regression. A multinomial XGBoost model was developed to classify phenotypes. Results Among 1048 patients, three inflammatory phenotypes (C1, C2, C3) were identified and validated in two independent cohorts. The phenotype C1, with a larger proportion of poorly/non-inflated lung compartments, had the highest 90-day mortality and shock incidence and fewest ventilator-free days, followed by C3, while C2 patients had the best outcomes (p<0.001). Phenotype C1 was characterised by intense innate immune activation, cytokine amplification and metabolic reprogramming. Phenotype C2 demonstrated immune suppression, ...