Therapeutic potential of pachymic acid in glioma: a comprehensive approach combining experimental validation, network pharmacology, and machine learning.

Pachymic acid (PA) is a natural active component of Poria cocos(Schw.)Wol. Although PA exhibits antitumor activity in multiple cancers, its effects and mechanisms against glioma remain elusive. This study aimed to investigate the anti-glioma effects of PA and identify its potential candidate core targets and pathways. Functional experiments demonstrated that PA significantly inhibited the proliferation, migration, and invasion of glioma cells and induced apoptosis in a dose-dependent manner. Using network pharmacology, machine learning, and bioinformatics analysis, we screened six candidate targets: PDE4D, CAPN2, MAPK9, EPHB6, CCKBR, and FAAH. TCGA analysis confirmed that CAPN2 and PDE4D were upregulated in glioma, while the others were downregulated. Molecular docking, an in silico predictive approach, suggested that PA may potentially form favorable binding conformations with these core targets based on calculated binding energies. Further verification revealed that PA treatment was associated with reduced AKT and mTOR phosphorylation, suggesting that the PI3K-AKT-mTOR pathway may be involved in mediating its anti-glioma activity. In conclusion, PA exerts dose-dependent anti-glioma cellular effects through multi-gene regulation, and the suppressed activation of the PI3K-AKT pathway may correlate with its anti-tumor function. These results provide a theoretical basis for developing PA as a potential therapeutic agent for glioma.