Scholarlink

Background <p>Ubiquinol-cytochrome c reductase hinge protein (UQCRH) is a component of mitochondrial respiratory chain complex CIII. Its relationship with human cancer has been less studied. Pathomics uses artificial intelligence algorithms to collect histopathological image features and perform joint analysis by combining gene and transcriptome data. In this study, a pathomics prediction model was established based on UQCRH expression and histopathological images of lung adenocarcinoma (LUAD). Prognostic value and other analyses were conducted based on this model.</p> Methods <p>The expression level of UQCRH in 33 types of human cancers was measured. Its relationship with the survival of the primary LUAD samples with complete pathological images, gene expression data, and clinical information were divided into high and low expression groups based on the expression level threshold of the UQCRH gene (Table <InternalRef RefID="MOESM1">S1</InternalRef>). LUAD patients was studied. Pathomic prediction model was established by using machine learning algorithms according to the UQCRH expression level and the characteristics of LUAD histopathological images. Based on this prediction model, survival analysis, molecular pathways, immune infiltration, immunological subtypes, ICI treatment prediction, and drug sensitivity analyses were performed.</p> Results <p>UQCRH is highly expressed in various cancers, including LUAD. In addition, we verified that UQCRH is overexpressed in human LUAD tissues. High expression of UQCRH is worse prognostic factor for LUAD patients. A pathomic prediction model was constructed based on the UQCRH expression level and histopathology image features. The pathomic score showed good correlation with the UQCRH expression level. Patients in the high-risk group of the pathomic prediction model had worse prognosis and higher tumor proliferation ability, but may have better response to immune checkpoint inhibitors (ICIs) therapy.</p> Conclusion <p>We have established a pathomic prediction model for LUAD based on gene expression values and according to histopathological image features, which can predict patient survival prognosis and has potential guiding value for ICIs therapy.</p> Graphical Abstract <p></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

Pathomic model to predict the expression of UQCRH and overall survival of lung adenocarcinoma patients

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月13日

查看全文
  • Yong Chen,
  • Jie Liu,
  • Guoping Li,
  • Huifang Huang

摘要

Background

Ubiquinol-cytochrome c reductase hinge protein (UQCRH) is a component of mitochondrial respiratory chain complex CIII. Its relationship with human cancer has been less studied. Pathomics uses artificial intelligence algorithms to collect histopathological image features and perform joint analysis by combining gene and transcriptome data. In this study, a pathomics prediction model was established based on UQCRH expression and histopathological images of lung adenocarcinoma (LUAD). Prognostic value and other analyses were conducted based on this model.

Methods

The expression level of UQCRH in 33 types of human cancers was measured. Its relationship with the survival of the primary LUAD samples with complete pathological images, gene expression data, and clinical information were divided into high and low expression groups based on the expression level threshold of the UQCRH gene (Table S1). LUAD patients was studied. Pathomic prediction model was established by using machine learning algorithms according to the UQCRH expression level and the characteristics of LUAD histopathological images. Based on this prediction model, survival analysis, molecular pathways, immune infiltration, immunological subtypes, ICI treatment prediction, and drug sensitivity analyses were performed.

Results

UQCRH is highly expressed in various cancers, including LUAD. In addition, we verified that UQCRH is overexpressed in human LUAD tissues. High expression of UQCRH is worse prognostic factor for LUAD patients. A pathomic prediction model was constructed based on the UQCRH expression level and histopathology image features. The pathomic score showed good correlation with the UQCRH expression level. Patients in the high-risk group of the pathomic prediction model had worse prognosis and higher tumor proliferation ability, but may have better response to immune checkpoint inhibitors (ICIs) therapy.

Conclusion

We have established a pathomic prediction model for LUAD based on gene expression values and according to histopathological image features, which can predict patient survival prognosis and has potential guiding value for ICIs therapy.

Graphical Abstract