Scholarlink

<p>Glutamine, a key component in cancer cell survival, has garnered interest as a potential target for cancer antimetabolic therapy. However, the clinical relevance of abnormal glutamine metabolism (GM) in colorectal cancer (CRC) patients remains largely unexplored. In this study, we collected 904 GM-related genes (GMGs), and devised a machine learning computational framework to develop a glutamine metabolism immunity index (GMII) for CRC patients. Through the utilization of median GMII scores, we stratified CRC patients into two categories: those with low-GMII and those with high-GMII. We found that the high-GMII patients had markedly inferior overall survival (OS) compared to their low-GMII patients in the TCGA-COAD cohort (<i>P</i> &lt; 0.001), and the AUC values of the GMII for OS were 0.749 at 1&#xa0;years, 0.751 at 3&#xa0;years, and 0.788 at 5&#xa0;years for the TCGA-COAD cohort, 0.579 at 1&#xa0;years, 0.559 at 3&#xa0;years, and 0.602 at 5&#xa0;years for the GSE17536 cohort, 0.655 at 1&#xa0;years, 0.604 at 3&#xa0;years, and 0.635 at 5&#xa0;years for the GSE17537 cohort. A nomogram was established by integrating clinical features with GMII, offering a precise and dependable tool for clinical management of CRC patients. C-index and decision curve analysis showed that the GMII-based nomogram outperformed other clinical models in terms of accuracy and net clinical benefit. Importantly, the low-GMII patients had better efficacy for conventional chemotherapeutic drugs (<i>P</i> &lt; 0.0001), suggesting that the GMII scoring system may guide chemotherapy drug selection in CRC patients. In conclusion, GMII represents a potent and promising tool for drug screening, clinical decision-making, prognosis prediction, and enhancing the efficacy of immunotherapy in CRC patients. It can facilitate the identification of patients who are likely to benefit from immunotherapy and enable the development of precise and personalized treatment strategies.</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

Machine learning develops a glutamine metabolism-related signature for predicting clinical outcome, molecular subtyping, immunotherapy efficacy and drug sensitivity in colorectal cancer

  • 期刊论文

  • 发表日期：

    2025年5月28日

查看全文
  • Chunhong Li,
  • Yuhua Mao,
  • Yi Liu,
  • Chunchun Su,
  • Jiahua Hu,
  • Haiyin Tan,
  • Xiao Fan,
  • Yiming Mao

摘要

Glutamine, a key component in cancer cell survival, has garnered interest as a potential target for cancer antimetabolic therapy. However, the clinical relevance of abnormal glutamine metabolism (GM) in colorectal cancer (CRC) patients remains largely unexplored. In this study, we collected 904 GM-related genes (GMGs), and devised a machine learning computational framework to develop a glutamine metabolism immunity index (GMII) for CRC patients. Through the utilization of median GMII scores, we stratified CRC patients into two categories: those with low-GMII and those with high-GMII. We found that the high-GMII patients had markedly inferior overall survival (OS) compared to their low-GMII patients in the TCGA-COAD cohort (P < 0.001), and the AUC values of the GMII for OS were 0.749 at 1 years, 0.751 at 3 years, and 0.788 at 5 years for the TCGA-COAD cohort, 0.579 at 1 years, 0.559 at 3 years, and 0.602 at 5 years for the GSE17536 cohort, 0.655 at 1 years, 0.604 at 3 years, and 0.635 at 5 years for the GSE17537 cohort. A nomogram was established by integrating clinical features with GMII, offering a precise and dependable tool for clinical management of CRC patients. C-index and decision curve analysis showed that the GMII-based nomogram outperformed other clinical models in terms of accuracy and net clinical benefit. Importantly, the low-GMII patients had better efficacy for conventional chemotherapeutic drugs (P < 0.0001), suggesting that the GMII scoring system may guide chemotherapy drug selection in CRC patients. In conclusion, GMII represents a potent and promising tool for drug screening, clinical decision-making, prognosis prediction, and enhancing the efficacy of immunotherapy in CRC patients. It can facilitate the identification of patients who are likely to benefit from immunotherapy and enable the development of precise and personalized treatment strategies.