Background: Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal cancer with limited treatment options. The PI3K/AKT/mTOR pathway is commonly activated in PDAC and plays a critical role in its progression.
Methods and results: In this study, the effect of taselisib (a selective PI3K inhibitor) on PDAC cell proliferation was investigated, and a significant decrease in viability was observed with increasing concentrations of taselisib. Differential analysis on samples from the Genotype-Tissue Expression and The Cancer Genome Atlas databases revealed 24 dysregulated PI3K/AKT/mTOR pathway-related genes (PRGs). Unsupervised clustering-based analysis of transcriptome cohorts revealed two clusters with high consistency between RNA-seq and microarray cohorts. Cluster B had higher enrichment of immune cells, particularly CD8+ T cells, and lower levels of immunosuppressive Treg cells. Moreover, we investigated the relationship between drug sensitivity and different clusters and found that cluster A had a better response to PI3K/AKT/mTOR pathway-related inhibitors and chemotherapy. Finally, cluster A exhibited significant activation of PI3K/AKT/mTOR and related oncogenic pathways, contributing to poor prognosis. The study also developed a risk score based on the expression profiles of PRGs and machine learning, which showed a significant increase in overall survival time among patients in the low-risk group. Importantly, the PI3K/AKT/mTOR pathway could be used to better predict individual risk scores, as evidenced by stratified survival analysis.
Conclusions: These findings suggest that targeting the PI3K/AKT/mTOR pathway may have therapeutic potential in PDAC, and distinct pathway states, immune modulation and tumor microenvironments have prognostic value.
Keywords: PI3K/AKT/mTOR; immune microenvironment; machine learning; pancreatic ductal adenocarcinoma; prognosis.
© 2023 John Wiley & Sons, Ltd.