Mining the plasma-proteome associated genes in patients with gastro-esophageal cancers for biomarker discovery

Frederick S. Vizeacoumar, Hongyu Guo, Lynn Dwernychuk, Adnan Zaidi, Andrew Freywald, Fang‐Xiang Wu, Franco J. Vizeacoumar, Shahid AhmedScientific 2021 · DOI

Abstract

Gastro-esophageal (GE) cancers are one of the major causes of cancer-related death in the world. There is a need for novel biomarkers in the management of GE cancers, to yield predictive response to the available therapies. Our study aims to identify leading genes that are differentially regulated in patients with these cancers. We explored the expression data for those genes whose protein products can be detected in the plasma using the Cancer Genome Atlas to identify leading genes that are differentially regulated in patients with GE cancers. Our work predicted several candidates as potential biomarkers for distinct stages of GE cancers, including previously identified CST1, INHBA, STMN1, whose expression correlated with cancer recurrence, or resistance to adjuvant therapies or surgery. To define the predictive accuracy of these genes as possible biomarkers, we constructed a co-expression network and performed complex network analysis to measure the importance of the genes in terms of a ratio of closeness centrality (RCC). Furthermore, to measure the significance of these differentially regulated genes, we constructed an SVM classifier using machine learning approach and verified these genes by using receiver operator characteristic (ROC) curve as an evaluation metric. The area under the curve measure was > 0.9 for both the overexpressed and downregulated genes suggesting the potential use and reliability of these candidates as biomarkers. In summary, we identified leading differentially expressed genes in GE cancers that can be detected in the plasma proteome. These genes have potential to become diagnostic and therapeutic biomarkers for early detection of cancer, recurrence following surgery and for development of targeted treatment.

Introduction

Gastro-esophageal (GE) cancers, encompassing stomach and esophageal malignancies, represent a significant global health burden due to their aggressive nature and high mortality rates. Current diagnostic methods often rely on invasive tissue biopsies, and existing biomarkers like HER2 are only applicable to a limited patient subset. There is a critical need for non-invasive "liquid biopsy" candidates—biomarkers detectable in the blood plasma—to facilitate early detection, monitor recurrence, and predict therapeutic responses.

Methodology

This study employed a multi-stage computational pipeline to bridge the gap between genomic research and clinical diagnostic potential. First, gene expression data from The Cancer Genome Atlas (TCGA)—specifically 450 stomach adenocarcinoma and 196 esophageal carcinoma samples—was filtered against the Plasma Proteome Database. This ensured the focus remained on ~1,200 genes whose protein products are known to circulate in human blood.

Differential expression analysis using non-parametric tests identified genes significantly up- or down-regulated in cancer tissues compared to normal tissues. To understand functional interactions, Weighted Gene Co-expression Networks (WGCNA) were constructed. A novel metric, the Ratio of Closeness Centrality (RCC), was introduced to rank genes based on their network influence rather than just expression levels. Finally, a Support Vector Machine (SVM) classifier was trained to evaluate the diagnostic accuracy of these leading gene signatures.

Results and Discussion

The analysis identified several high-confidence biomarkers. For esophageal cancer, top upregulated genes included BIRC5, APOC2, CENPF, STMN1, and HNRPC. Stomach cancer signatures were prominent for CST1, INHBA, ACAN, HSP90AB1, and HSPD1.

Expression of BIRC5 across different cancer stages compared to normal (N) tissue.

As shown in the figure above, BIRC5 (survivin) exhibits significant upregulation across all clinical stages compared to normal tissue, suggesting its utility in both early detection and monitoring. Similarly, STMN1 showed a stark contrast between tumor and normal samples, reinforcing its potential as a robust diagnostic marker.

Differential expression of STMN1 in normal (N) vs. tumor (T) samples.

The SVM classifier achieved remarkable performance, with an Area Under the Curve (AUC) of 0.9941 for esophageal cancer and 0.9924 for stomach cancer. This indicates that the identified plasma-proteome-associated genes can distinguish cancer patients from healthy controls with near-perfect accuracy. The study also identified stage-specific signatures, including genes that increase progressively as cancer advances, which is critical for clinical staging.

Conclusion

By intersecting genomic data with plasma proteome accessibility, this research identifies a robust signature of blood-detectable biomarkers for GE cancers. While these computational results are mathematically impressive, the authors acknowledge limitations such as small control groups in the TCGA datasets and the lack of histopathological stratification. Future work must involve prospective clinical trials to measure these protein markers directly in patient blood samples to confirm their real-world diagnostic and predictive utility.

Summary updated

Abstract

Gastro-esophageal (GE) cancers are one of the major causes of cancer-related death in the world. There is a need for novel biomarkers in the management of GE cancers, to yield predictive response to the available therapies. Our study aims to identify leading genes that are differentially regulated in patients with these cancers. We explored the expression data for those genes whose protein products can be detected in the plasma using the Cancer Genome Atlas to identify leading genes that are differentially regulated in patients with GE cancers. Our work predicted several candidates as potential biomarkers for distinct stages of GE cancers, including previously identified CST1, INHBA, STMN1, whose expression correlated with cancer recurrence, or resistance to adjuvant therapies or surgery. To define the predictive accuracy of these genes as possible biomarkers, we constructed a co-expression network and performed complex network analysis to measure the importance of the genes in terms of a ratio of closeness centrality (RCC). Furthermore, to measure the significance of these differentially regulated genes, we constructed an SVM classifier using machine learning approach and verified these genes by using receiver operator characteristic (ROC) curve as an evaluation metric. The area under the curve measure was > 0.9 for both the overexpressed and downregulated genes suggesting the potential use and reliability of these candidates as biomarkers. In summary, we identified leading differentially expressed genes in GE cancers that can be detected in the plasma proteome. These genes have potential to become diagnostic and therapeutic biomarkers for early detection of cancer, recurrence following surgery and for development of targeted treatment.
BibTeX
@inproceedings{vizeacoumar2021mining,
  author = {Frederick S. Vizeacoumar and Hongyu Guo and Lynn Dwernychuk and Adnan Zaidi and Andrew Freywald and Fang‐Xiang Wu and Franco J. Vizeacoumar and Shahid Ahmed},
  title = {{Mining the plasma-proteome associated genes in patients with gastro-esophageal cancers for biomarker discovery}},
  booktitle = {Scientific},
  year = {2021},
  doi = {10.1038/s41598-021-87037-w},
}
Figure