Wang Jingyuan, Yan Shuai
Department of Neurological Surgery, The First Affiliated Hospital of China Medical University, Shenyang, China.
Department of Neurological Function Examination, Affiliated Hospital of Hebei University, Baoding, China.
Front Pharmacol. 2025 Jan 13;15:1523779. doi: 10.3389/fphar.2024.1523779. eCollection 2024.
Lower-grade glioma (LGG) exhibits significant heterogeneity in clinical outcomes, and current prognostic markers have limited predictive value. Despite the growing recognition of histone modifications in tumor progression, their role in LGG remains poorly understood. This study aimed to develop a histone modification-based risk signature and investigate its relationship with drug sensitivity to guide personalized treatment strategies.
We performed single-cell RNA sequencing analysis on LGG samples (n = 4) to characterize histone modification patterns. Through integrative analysis of TCGA-LGG (n = 513) and CGGA datasets (n = 693 and n = 325), we constructed a histone modification-related risk signature (HMRS) using machine learning approaches. The model's performance was validated in multiple independent cohorts. We further conducted comprehensive analyses of molecular mechanisms, immune microenvironment, and drug sensitivity associated with the risk stratification.
We identified distinct histone modification patterns across five major cell populations in LGG and developed a robust 20-gene HMRS from 129 candidate genes that effectively stratified patients into high- and low-risk groups with significantly different survival outcomes (training set: AUC = 0.77, 0.73, and 0.71 for 1-, 3-, and 5-year survival; < 0.001). Integration of HMRS with clinical features further improved prognostic accuracy (C-index >0.70). High-risk tumors showed activation of TGF-β and IL6-JAK-STAT3 signaling pathways, and distinct mutation profiles including TP53 (63% vs 28%), IDH1 (68% vs 85%), and ATRX (46% vs 20%) mutations. The high-risk group demonstrated significantly elevated immune and stromal scores ( < 0.001), with distinct patterns of immune cell infiltration, particularly in memory CD4 T cells ( < 0.001) and CD8 T cells ( = 0.001). Drug sensitivity analysis revealed significant differential responses to six therapeutic agents including Temozolomide and targeted drugs ( < 0.05).
Our study establishes a novel histone modification-based prognostic model that not only accurately predicts LGG patient outcomes but also reveals potential therapeutic targets. The identified associations between risk stratification and drug sensitivity provide valuable insights for personalized treatment strategies. This integrated approach offers a promising framework for improving LGG patient care through molecular-based risk assessment and treatment selection.
低级别胶质瘤（LGG）在临床预后方面表现出显著的异质性，目前的预后标志物预测价值有限。尽管人们越来越认识到组蛋白修饰在肿瘤进展中的作用，但其在LGG中的作用仍知之甚少。本研究旨在开发一种基于组蛋白修饰的风险特征，并研究其与药物敏感性的关系，以指导个性化治疗策略。
我们对LGG样本（n = 4）进行了单细胞RNA测序分析，以表征组蛋白修饰模式。通过对TCGA-LGG（n = 513）和CGGA数据集（n = 693和n = 325）的综合分析，我们使用机器学习方法构建了一个与组蛋白修饰相关的风险特征（HMRS）。该模型的性能在多个独立队列中得到验证。我们进一步对与风险分层相关的分子机制、免疫微环境和药物敏感性进行了全面分析。
我们在LGG的五个主要细胞群体中确定了不同的组蛋白修饰模式，并从129个候选基因中开发了一个强大的20基因HMRS，该特征有效地将患者分为高风险和低风险组，其生存结果有显著差异（训练集：1年、3年和5年生存率的AUC分别为0.77、0.73和0.71；P < 0.001）。将HMRS与临床特征相结合进一步提高了预后准确性（C指数>0.70）。高风险肿瘤显示TGF-β和IL6-JAK-STAT3信号通路激活，以及不同的突变谱，包括TP53（63%对28%）、IDH1（68%对85%）和ATRX（46%对20%）突变。高风险组的免疫和基质评分显著升高（P < 0.001），免疫细胞浸润模式不同，特别是在记忆性CD4 T细胞（P < 0.001）和CD8 T细胞（P = 0.001）中。药物敏感性分析显示对包括替莫唑胺和靶向药物在内的六种治疗药物有显著的差异反应（P < 0.05）。
我们的研究建立了一种基于组蛋白修饰的新型预后模型，该模型不仅能准确预测LGG患者的预后，还能揭示潜在的治疗靶点。所确定的风险分层与药物敏感性之间的关联为个性化治疗策略提供了有价值的见解。这种综合方法为通过基于分子的风险评估和治疗选择改善LGG患者护理提供了一个有前景的框架。
