Research Paper Volume 12, Issue 19 pp 19159—19172
Transcriptomic analysis reveals a WNT signaling pathway-based gene signature prognostic for non-small cell carcinoma
- 1 Institute of Biological Sciences, Fudan University, Shanghai, P.R. China
- 2 Shanghai Key Laboratory of Molecular Imaging, Shanghai University of Medicine and Health Sciences, Shanghai, P.R. China
- 3 Department of Nuclear Medicine, Shanghai Chest Hospital, Shanghai Jiaotong University, Shanghai, P.R. China
Received: May 25, 2020 Accepted: July 3, 2020 Published: October 7, 2020https://doi.org/10.18632/aging.103724
How to Cite
Copyright: © 2020 Liu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
The value of combining multiple candidate genes into a panel to improve biomarker performance is increasingly emphasized. Genes associated with WNT signaling are widely-reported to provide prognostic signatures in non-small cell carcinoma (NSCLC). Screening of genes involved in this signaling pathway facilitated selection of an optimal candidate biomarker gene combination and development of an NSCLC prognostic model based on expression of these genes. Risk scores derived from the model performed well in predicting survival; in the training dataset, samples achieving a high risk score exhibit a shorter survival interval (median survival time 34.8 months, 95% CI 31.1-41.0) than did samples achieving a low risk score (median survival time 72.0 months, 95% CI 59.3-87.5, p=2e-11), and exhibited higher oncogene and lower tumor suppressor gene expression. Receiver-operator characteristic curves based on three-year survival demonstrate that the model outperformed clinical prognostic indicators. In addition, the model was validated in four independent cohorts, demonstrating robust NSCLC prognostic value. Correlation analyses reveal that the model offers efficacy independent of other clinical indicators. Gene Set Enrichment Analysis (GSEA) reveals that the model reflects variable tissue functional states relevant to NSCLC biology. In summary, the signature model shows potential as a valuable and robust NSCLC prognostic indicator.