Open Access

Machine learning‑based construction of damage‑associated molecular patterns related score identifies subtypes of pancreatic adenocarcinoma with distinct prognosis

  • Authors:
    • Jing Liang
    • Hui Wu
    • Zewen Song
    • Guoyin Li
    • Jianfeng Zhang
    • Wenxin Ding
  • View Affiliations

  • Published online on: March 24, 2025     https://doi.org/10.3892/ol.2025.14992
  • Article Number: 246
  • Copyright: © Liang et al. This is an open access article distributed under the terms of Creative Commons Attribution License.

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )
Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )


Abstract

The present study aimed to assess the prognostic significance of Damage‑Associated Molecular Pattern (DAMP)‑related gene expression in pancreatic adenocarcinoma (PAAD) and to develop a scoring system based on these genes. Consensus clustering was performed on patients with PAAD using data from The Cancer Genome Atlas (TCGA) and Meta‑cohort datasets, identifying three distinct clusters: C1 (pro‑DAMP), C2 (intermediate) and C3 (anti‑DAMP). Differential gene expression analysis between clusters C1 and C3 identified 141 significant genes. Least Absolute Shrinkage and Selection Operator Cox regression was utilized to derive an optimal predictor set, leading to the identification of six hub genes associated with the DAMP status, which were then employed to calculate the DAMPscore. Weighted Gene Co‑expression Network Analysis revealed a strong correlation between these eight hub genes and the DAMPscore. The functionality of these hub genes in PAAD was validated using a Cell Counting Kit‑8 assay and Transwell assays. The results indicated that patients with PAAD with elevated DAMPscores exhibited significantly reduced survival times. Receiver operating characteristic (ROC) curve analysis indicated that the DAMPscore has robust prognostic capabilities. In the Meta‑cohort, the area under the ROC curve (AUC) values for the DAMPscore to predict overall survival at 1, 3 and 5 years were 0.65, 0.70 and 0.77, respectively, while the AUC values for the TCGA‑PAAD cohort were 0.71, 0.73 and 0.72, respectively. Additional cohorts, such as E‑MTAB‑6134 and ICGC‑AU, corroborated the predictive power of the DAMPscore. A comparison of the DAMPscore with other prognostic models revealed that it consistently exhibited a superior C‑index across most PAAD cohorts. Furthermore, in vitro experiments demonstrated that PLEK2, a hub gene related to the DAMPscore, is involved in critical biological processes such as cell proliferation, migration and invasion. In conclusion, the DAMPscore is a promising prognostic biomarker for PAAD, surpassing traditional models in various datasets. This study emphasizes the role of DAMP‑related pathways in influencing tumor biology and highlights the importance of immune modulation in PAAD prognosis, suggesting that therapeutic strategies targeting DAMP signaling could improve patient outcomes.

Related Articles

Journal Cover

May-2025
Volume 29 Issue 5

Print ISSN: 1792-1074
Online ISSN:1792-1082

Sign up for eToc alerts

Recommend to Library

Copy and paste a formatted citation
x
Spandidos Publications style
Liang J, Wu H, Song Z, Li G, Zhang J and Ding W: Machine learning‑based construction of damage‑associated molecular patterns related score identifies subtypes of pancreatic adenocarcinoma with distinct prognosis. Oncol Lett 29: 246, 2025.
APA
Liang, J., Wu, H., Song, Z., Li, G., Zhang, J., & Ding, W. (2025). Machine learning‑based construction of damage‑associated molecular patterns related score identifies subtypes of pancreatic adenocarcinoma with distinct prognosis. Oncology Letters, 29, 246. https://doi.org/10.3892/ol.2025.14992
MLA
Liang, J., Wu, H., Song, Z., Li, G., Zhang, J., Ding, W."Machine learning‑based construction of damage‑associated molecular patterns related score identifies subtypes of pancreatic adenocarcinoma with distinct prognosis". Oncology Letters 29.5 (2025): 246.
Chicago
Liang, J., Wu, H., Song, Z., Li, G., Zhang, J., Ding, W."Machine learning‑based construction of damage‑associated molecular patterns related score identifies subtypes of pancreatic adenocarcinoma with distinct prognosis". Oncology Letters 29, no. 5 (2025): 246. https://doi.org/10.3892/ol.2025.14992