Open Access

ARTICLE

Using Multi-Omics Analysis to Explore Diagnostic Tool and Optimize Drug Therapy Selection for Patients with Glioma Based on Cross-Talk Gene Signature

by YUSHI YANG^1,#, CHUJIAO HU^2,#, SHAN LEI³, XIN BAO³, ZHIRUI ZENG^3,*, WENPENG CAO^4,*

1 Department of Pathology and Pathophysiology, School of Basic Medicine, Guizhou Medical University, Guiyang, 550025, China
2 Guizhou Provincial Engineering Technology Research Center for Chemical Drug R&D; State Key Laboratory of Functions and Applications of Medicinal Plants, Guizhou Medical University, Guiyang, 550014, China
3 Department of Physiology, School of Basic Medicine, Guizhou Medical University, Guiyang, 550025, China
4 Department of Anatomy, School of Basic Medicine, Guizhou Medical University, Key Laboratory of Human Brain Bank for Functions and Diseases of Department of Education of Guizhou Province, Guizhou Medical University, Guiyang, 550025, China

* Corresponding Authors: ZHIRUI ZENG. Email: ; WENPENG CAO. Email:
# Equal contribution

(This article belongs to the Special Issue: Multi-Omics Approaches for Precision Medicine)

Oncology Research 2024, 32(12), 1921-1934. https://doi.org/10.32604/or.2024.046191

Received 21 September 2023; Accepted 28 February 2024; Issue published 13 November 2024

Abstract

Background: The heterogeneity of prognosis and treatment benefits among patients with gliomas is due to tumor microenvironment characteristics. However, biomarkers that reflect microenvironmental characteristics and predict the prognosis of gliomas are limited. Therefore, we aimed to develop a model that can effectively predict prognosis, differentiate microenvironment signatures, and optimize drug selection for patients with glioma. Materials and Methods: The CIBERSORT algorithm, bulk sequencing analysis, and single-cell RNA (scRNA) analysis were employed to identify significant cross-talk genes between M2 macrophages and cancer cells in glioma tissues. A predictive model was constructed based on cross-talk gene expression, and its effect on prognosis, recurrence prediction, and microenvironment characteristics was validated in multiple cohorts. The effect of the predictive model on drug selection was evaluated using the OncoPredict algorithm and relevant cellular biology experiments. Results: A high abundance of M2 macrophages in glioma tissues indicates poor prognosis, and cross-talk between macrophages and cancer cells plays a crucial role in shaping the tumor microenvironment. Eight genes involved in the cross-talk between macrophages and cancer cells were identified. Among them, periostin (POSTN), chitinase 3 like 1 (CHI3L1), serum amyloid A1 (SAA1), and matrix metallopeptidase 9 (MMP9) were selected to construct a predictive model. The developed model demonstrated significant efficacy in distinguishing patient prognosis, recurrent cases, and characteristics of high inflammation, hypoxia, and immunosuppression. Furthermore, this model can serve as a valuable tool for guiding the use of trametinib. Conclusions: In summary, this study provides a comprehensive understanding of the interplay between M2 macrophages and cancer cells in glioma; utilizes a cross-talk gene signature to develop a predictive model that can predict the differentiation of patient prognosis, recurrence instances, and microenvironment characteristics; and aids in optimizing the application of trametinib in glioma patients.

---

Keywords

---

Supplementary Material

Supplementary Material File

---