Scholarlink

<p>In the field of gastrointestinal malignancies, colorectal cancer (CRC) is a globally prevalent and lethal disease. An increasing amount of research indicates that ferroptosis plays a role in various aspects of malignant tumor progression, indicating its considerable potential for cancer diagnosis and therapy. For this research, we extracted 564 ferroptosis-related genes (FRGs) from the FerrDb V2 database, and devised a machine learning computational framework to develop a novel ferroptosis-related survival prognostic model for CRC patients. A riskScore was calculated for each CRC patient, and the CRC patients were divided into two groups: those at low-risk group and those at high-risk group. The model comprises seven differentially expressed ferroptosis-related genes (DRD4, HOTAIR, ASNS, GSTM1, TFAP2C, PANX2, and HAMP), which can function as standalone prognostic markers for CRC patients. Analysis of survival data revealed that in the TCGA-COAD training cohort, patients classified within the high-risk group showed a notably poorer outcome compared to those in the low-risk group (<i>P</i> &lt; 0.001), and a similar phenomenon was observed in the validation cohorts, including the GSE17536 cohort (<i>P</i> = 0.032) and GSE29621 cohort (<i>P</i> = 0.011). Additionally, we observed differences in immune infiltration and immune function among the different risk groups. Notably, the high-risk group demonstrated reduced sensitivity to most commonly used chemotherapeutic drugs. Therefore, this molecular diagnostic model is proposed as a means of assessing the prognostic risk of CRC patients, providing new perspectives and potential candidates for immune-enhancing therapies for CRC.</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 machine learning computational framework develops a novel ferroptosis-related signature for improving clinical outcome in colorectal cancer

  • 期刊论文

  • 发表日期：

    2026年3月14日

查看全文
  • Chunhong Li,
  • Yuhua Mao,
  • Xiaoping Pang,
  • Jiahua Hu,
  • Xianliang Hou,
  • Minglin Ou

摘要

In the field of gastrointestinal malignancies, colorectal cancer (CRC) is a globally prevalent and lethal disease. An increasing amount of research indicates that ferroptosis plays a role in various aspects of malignant tumor progression, indicating its considerable potential for cancer diagnosis and therapy. For this research, we extracted 564 ferroptosis-related genes (FRGs) from the FerrDb V2 database, and devised a machine learning computational framework to develop a novel ferroptosis-related survival prognostic model for CRC patients. A riskScore was calculated for each CRC patient, and the CRC patients were divided into two groups: those at low-risk group and those at high-risk group. The model comprises seven differentially expressed ferroptosis-related genes (DRD4, HOTAIR, ASNS, GSTM1, TFAP2C, PANX2, and HAMP), which can function as standalone prognostic markers for CRC patients. Analysis of survival data revealed that in the TCGA-COAD training cohort, patients classified within the high-risk group showed a notably poorer outcome compared to those in the low-risk group (P < 0.001), and a similar phenomenon was observed in the validation cohorts, including the GSE17536 cohort (P = 0.032) and GSE29621 cohort (P = 0.011). Additionally, we observed differences in immune infiltration and immune function among the different risk groups. Notably, the high-risk group demonstrated reduced sensitivity to most commonly used chemotherapeutic drugs. Therefore, this molecular diagnostic model is proposed as a means of assessing the prognostic risk of CRC patients, providing new perspectives and potential candidates for immune-enhancing therapies for CRC.