Sun Liangdong, Wang Jue, Yu Huansha, Zhu Xinsheng, Zhang Jing, Hu Junjie, Yan Yilv, Zhang Xun, Zhu Yuming, Jiang Gening, Ding Ming, Zhang Peng, Zhang Lele
Department of Thoracic Surgery, Shanghai Pulmonary Hospital, School of Medicine, Tongji University, Shanghai, China.
Department of Animal Experimental Center, Shanghai Pulmonary Hospital, School of Medicine, Tongji University, Shanghai, China.
Commun Biol. 2025 Feb 1;8(1):152. doi: 10.1038/s42003-025-07595-x.
Lung squamous cell carcinoma (LUSC) represents a major subtype of lung cancer, and it demonstrates limited treatment options and worse survival. Identifications of a prognostic model and chemoresistance mechanism can be helpful for improving stratification and guiding therapy decisions. The integrative development of machine learning-based models reveals a random survival forest (RSF) prognostic model for LUSC. The 12-gene RSF model exhibits high prognostic power in more than 1,000 LUSC patients. High-risk LUSC patients are associated with worse survival and the activation of the epithelial-mesenchymal transition pathway. Additionally, high-risk LUSC patients are resistant to docetaxel or vinorelbine treatment. In vitro and in vivo drug sensitivity experiments indicates that high-risk HCC15/H226 tumour cells and cell line-derived xenograft models are more resistant to vinorelbine treatment. Furthermore, the combination of chemotherapy with transforming growth factor-β inhibition augments antitumour responses in LUSC tumours. Our study provides valuable insights into prognosis stratification and the development of therapeutic strategies for LUSC.
肺鳞状细胞癌（LUSC）是肺癌的一种主要亚型，其治疗选择有限，生存率较低。识别预后模型和化疗耐药机制有助于改善分层并指导治疗决策。基于机器学习模型的综合开发揭示了一种用于LUSC的随机生存森林（RSF）预后模型。该12基因RSF模型在1000多名LUSC患者中表现出较高的预后能力。高危LUSC患者的生存率较差，且上皮-间质转化途径被激活。此外，高危LUSC患者对多西他赛或长春瑞滨治疗耐药。体外和体内药物敏感性实验表明，高危HCC15/H226肿瘤细胞和细胞系衍生的异种移植模型对长春瑞滨治疗更耐药。此外，化疗与转化生长因子-β抑制相结合可增强LUSC肿瘤的抗肿瘤反应。我们的研究为LUSC的预后分层和治疗策略的制定提供了有价值的见解。
