Machine learning identified promising markers for cervical cancer prognosis and immune response from sphingolipid metabolism
New Medicine · 10.03.2025 · 3 (1) · 82-101
DOI: 10.61958/nmog9547 ↗Kurz gefasst
Fachartikel in der Zeitschrift New Medicine, online erschienen am 10.03.2025. Verfasst von Jun-Jie Zhu und 4 weiteren Autor:innen. Die Arbeit nennt 61 Literaturangaben.
Abstract (Originalsprache)
Background
Cervical cancer (CC) poses a significant women's health challenge, with limited therapeutic efficacy for advanced cases. Recent studies suggest sphingolipid metabolism plays a critical role in tumor progression and immune regulation, offering potential prognostic markers and therapeutic targets.
Methods
We employed bioinformatics and machine learning approaches to analyze CC datasets from TCGA and GEO. Differentially expressed sphingolipid metabolism-related genes (DESMRGs) were identified, followed by LASSO regression and an optimized XGBoost algorithm to construct a prognostic model. Immune profiling, functional enrichment, and drug sensitivity analyses were performed to explore molecular mechanisms and therapeutic implications.
Results
Four key DESMRGs (DEGS2, LYN, CERS4, ALOX12B) were identified as prognostic markers, forming a robust model with high predictive accuracy (AUC: 0.906-0.948). The model stratified patients into high- and low-risk groups with significant survival differences. Functional enrichment revealed these genes' roles in lipid metabolism and immune modulation. Immune profiling indicated enhanced dendritic cell infiltration in the low-risk group, while high-risk patients showed elevated myocyte levels. Drug sensitivity analysis highlighted potential therapeutic agents tailored to risk groups.
Conclusions
This study developed a novel prognostic model based on sphingolipid metabolism-related genes, providing insights into CC biology and informing personalized treatment strategies.
Zitieren
APA
Zhu, J., Wang, X., Liang, Y., Zhou, W., & Guo, C. (2025). Machine learning identified promising markers for cervical cancer prognosis and immune response from sphingolipid metabolism. New Medicine, 3(1), 82-101. https://doi.org/10.61958/nmog9547
Vancouver
Zhu J, Wang X, Liang Y, Zhou W, Guo C. Machine learning identified promising markers for cervical cancer prognosis and immune response from sphingolipid metabolism. New Medicine. 2025;3(1):82-101. doi:10.61958/nmog9547
BibTeX
@article{Zhu2025,
author = {Zhu, Jun-Jie and Wang, Xiao-Hai and Liang, Yu and Zhou, Wei and Guo, Chun-Ming},
title = {Machine learning identified promising markers for cervical cancer prognosis and immune response from sphingolipid metabolism},
journal = {New Medicine},
year = {2025},
volume = {3},
number = {1},
pages = {82--101},
doi = {10.61958/nmog9547}
}
EUVIMED