EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Integrative Single‐Cell and Machine Learning Analysis Identifies an EMT ‐Associated Prognostic Signature for Papillary Thyroid Cancer

Tianfeng Xu iD, Ruonan Sun, Yujie Zhang iD, Xun Zheng iD

Cancer Medicine · 2026

DOI: 10.1002/cam4.71766 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cancer Medicine, 2026. Verfasst von Tianfeng Xu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Division of Thyroid Surgery, Department of General Surgery, West China Hospital, Sichuan University, Chengdu, China. Hauptthemen laut MEDLINE: Epithelial-Mesenchymal Transition, Machine Learning, Thyroid Cancer, Papillary, Biomarkers, Tumor. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 92 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: The "Qimingxing" Research Fund for Young Talents of West China Hospital.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Epithelial–mesenchymal transition (EMT) plays a critical role in tumor progression; however, the underlying molecular mechanisms of EMT in papillary thyroid carcinoma (PTC) remain incompletely understood. This study aimed to investigate EMT‐related mechanisms in PTC using an integrative approach combining single‐cell RNA sequencing and machine learning.

Methods

Differentially expressed genes (DEGs) between PTC and normal thyroid tissues were identified, and EMT‐related candidate genes were obtained by intersecting DEGs with EMT‐related genes (EMT‐RGs). Prognostic genes were screened using univariate Cox regression, and a risk model was constructed based on 101 machine learning algorithm combinations. Patients were stratified into high‐ and low‐risk groups (HRG and LRG) according to risk scores, and the model was validated in an internal cohort. Additional analyses included nomogram construction, immune infiltration profiling, tumor mutational burden (TMB) assessment, drug sensitivity prediction, and molecular regulatory network analysis. Prognostic gene expression was further validated in vitro.

Results

Eight EMT‐related prognostic genes (TYRO3, E2F1, TNFSF15, TGFBR3, PTX3, FHL2, SNAI1, and WT1) were identified. Patients in the HRG exhibited significantly poorer overall survival than those in the LRG. The nomogram showed good predictive accuracy for survival estimation. Immune infiltration analysis revealed significant differences between risk groups across six immune‐related features. Splice site–related mutations were predominantly observed in the LRG but were absent in the HRG. Drug sensitivity analysis indicated higher sensitivity to BIRB.0796 in the LRG, whereas ABT‐263, AG‐014699, BX‐795, and DMOG were more effective in the HRG. Single‐cell analysis identified fibroblasts as key cell populations, with FHL2, PTX3, and TGFBR3 showing increased activity during critical differentiation stages. In vitro experiments confirmed expression patterns consistent with bioinformatics findings.

Conclusion

This study identifies eight EMT‐related prognostic genes in PTC and highlights their potential value as biomarkers for prognostic evaluation and therapeutic stratification.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Epithelial-Mesenchymal TransitionMachine LearningThyroid Cancer, PapillaryBiomarkers, TumorThyroid NeoplasmsSingle-Cell AnalysisHumansPrognosisGene Expression Regulation, NeoplasticFemaleGene Expression ProfilingMaleNomograms

Stichwörter der Autor:innen

101 machine learning algorithm combinationsepithelial mesenchymal transitionpapillary thyroid cancerprognostic genessingle‐cell RNA sequencing analysis

Substanzen

Biomarkers, Tumor

Autor:innen und Institutionen

  • Division of Thyroid Surgery, Department of General Surgery, West China Hospital, Sichuan University, Chengdu, China.
  • Precision Medicine Center, West China Hospital, Sichuan University, Chengdu, China.
  • Department of Ultrasonography, West China Hospital, Sichuan University, Chengdu, China.

ORCID-Profile: Tianfeng Xu, Yujie Zhang, Xun Zheng

Förderung

  • The "Qimingxing" Research Fund for Young Talents of West China Hospital

Zitieren

APA

Xu, T., Sun, R., Zhang, Y., & Zheng, X. (2026). Integrative Single‐Cell and Machine Learning Analysis Identifies an EMT ‐Associated Prognostic Signature for Papillary Thyroid Cancer. Cancer Medicine. https://doi.org/10.1002/cam4.71766

Vancouver

Xu T, Sun R, Zhang Y, Zheng X. Integrative Single‐Cell and Machine Learning Analysis Identifies an EMT ‐Associated Prognostic Signature for Papillary Thyroid Cancer. Cancer Medicine. 2026. doi:10.1002/cam4.71766

BibTeX
@article{Xu2026,
  author = {Xu, Tianfeng and Sun, Ruonan and Zhang, Yujie and Zheng, Xun},
  title = {Integrative Single‐Cell and Machine Learning Analysis Identifies an EMT ‐Associated Prognostic Signature for Papillary Thyroid Cancer},
  journal = {Cancer Medicine},
  year = {2026},
  doi = {10.1002/cam4.71766}
}
Kostenlos starten →