Scholarlink

<p>Pancreatic ductal adenocarcinoma (PDAC) ranks among the most lethal malignant tumours, characterised by an immunosuppressive tumour microenvironment and resistance to conventional therapies. Increasing evidence indicates that mitochondrial genes correlate with tumour progression, immune evasion, and treatment response. This study integrated transcriptomic data from multiple databases (GEO, TCGA, ICGC) to identify mitochondrial-related differentially expressed genes (MRGs) between tumour and normal tissues. Concurrently, a prognostic model for mitochondrial genes was constructed using 100 machine learning methods. The risk score from the optimal model, termed the MRGs score, effectively stratified patients into high-risk and low-risk groups. The model demonstrated robust predictive performance across training and validation cohorts, with patients in the high MRG score group exhibiting significantly poorer overall survival. Functional analysis revealed strong associations between MRG scores and cellular processes, including cell cycle regulation, immune cell infiltration, and metabolism. Computational deconvolution analysis revealed that high MRG scores correlate with increased infiltration of immunosuppressive cells and altered immune checkpoint expression. The existing MRGs score has the important property of predicting prognosis while at the same time capturing tumour microenvironment heterogeneity, genomic instability, and computationally predicted chemotherapy response variability, thus providing genuinely new avenues for developing risk stratification hypotheses and designing personalised treatment strategies for pancreatic cancer patients. However, the validity of these results must be confirmed by future clinical and laboratory studies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel mitochondrial-associated risk score reveals a potential link between pancreatic ductal adenocarcinoma and chemotherapy resistance as well as poor response to immunotherapy

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月22日

查看全文
  • Qi Zhang,
  • Guang Tan

摘要

Pancreatic ductal adenocarcinoma (PDAC) ranks among the most lethal malignant tumours, characterised by an immunosuppressive tumour microenvironment and resistance to conventional therapies. Increasing evidence indicates that mitochondrial genes correlate with tumour progression, immune evasion, and treatment response. This study integrated transcriptomic data from multiple databases (GEO, TCGA, ICGC) to identify mitochondrial-related differentially expressed genes (MRGs) between tumour and normal tissues. Concurrently, a prognostic model for mitochondrial genes was constructed using 100 machine learning methods. The risk score from the optimal model, termed the MRGs score, effectively stratified patients into high-risk and low-risk groups. The model demonstrated robust predictive performance across training and validation cohorts, with patients in the high MRG score group exhibiting significantly poorer overall survival. Functional analysis revealed strong associations between MRG scores and cellular processes, including cell cycle regulation, immune cell infiltration, and metabolism. Computational deconvolution analysis revealed that high MRG scores correlate with increased infiltration of immunosuppressive cells and altered immune checkpoint expression. The existing MRGs score has the important property of predicting prognosis while at the same time capturing tumour microenvironment heterogeneity, genomic instability, and computationally predicted chemotherapy response variability, thus providing genuinely new avenues for developing risk stratification hypotheses and designing personalised treatment strategies for pancreatic cancer patients. However, the validity of these results must be confirmed by future clinical and laboratory studies.