The role of RAB GTPases in predicting prognosis and therapy response in pancreatic cancer.

backgroundThe Rab family of small GTPases (RABs) regulates membrane trafficking and is implicated in tumor progression. Pancreatic cancer (PC), a highly aggressive malignancy, lacks effective therapeutic strategies. Using bioinformatics and experimental validation, we identified five RABs (RAB10, RAB11A, RAB39B, RAB28, and RAB11B) to construct a RAB prognostic model (RPM) for PC.

methodsDatasets from TCGA and GEO were analyzed using Cox and LASSO regression to establish RPM. A prognostic nomogram integrating RPM with clinical features was developed. Microenvironment and drug sensitivity analysis were performed to predict the relationship between RPM score and therapy response. In vitro experiments (CCK8, MDA assays, lentiviral knockdown, qPCR, and western blot) were performed to validate RAB39B’s role in autophagy-dependent ferroptosis.

resultsRPM stratified patients into high- and low-risk groups, with high-risk patients showing poorer overall survival. The low-risk group exhibited enriched immune infiltration, elevated HLA expression, higher dysfunctional T-cell scores, and lower tumor mutation burden, suggesting an exhausted tumor immune microenvironment which could benefit from immunotherapy. The high-risk cohort exhibited markedly reduced responsiveness to axitinib, gefitinib, imatinib, methotrexate, nilotinib, and sunitinib, while demonstrating enhanced sensitivity to erlotinib, lapatinib, and sorafenib. Correlation analysis further identified RAB39B as a critical modulator of treatment response. In vitro experiments demonstrated that RAB39B knockdown reduced ferroptosis inducibility and suppressed LC3-II accumulation, implicating RAB39B’s role in autophagy-dependent ferroptosis.

conclusionsRPM serves as a robust prognostic tool for PC. RAB39B drives autophagy-dependent ferroptosis and influences drug sensitivity, offering therapeutic insights for PC.