Gastrointestinal stromal tumors (GIST), the most common sarcomas of the gastrointestinal tract, are primarily driven by c-KIT or PDGFRA mutations that activate downstream signaling pathways, including PI3K/AKT/mTOR. Whereas imatinib (IM), a first-line tyrosine kinase inhibitor (TKI), is initially effective, resistance develops in ∼50% of patients within 20 months. Second- and third-line TKIs, such as sunitinib and regorafenib, provide limited benefits, highlighting the urgent need to address resistance mechanisms. Previous research has predominantly focused on genetic drivers of resistance and overlooked the role of the tumor microenvironment. In this study, we identified a role for cancer-associated fibroblasts (CAF) in driving IM resistance. Specifically, TGFβ1 secreted by CAFs amplified cellular communication network factor 2 (CCN2)/receptor for activated C kinase 1 (Rack1) signaling. The CCN2/Rack1 axis activated PI3K/AKT signaling to induce phosphorylation and mitochondrial translation of phosphoglycerate kinase 1 (PGK1), promoting metabolic reprogramming that supported tumor survival and drug resistance. Coculture models and single-cell RNA sequencing revealed distinct CAF subtypes and showed that CAF-secreted TGFβ1 enhanced glycolysis and inhibited the tricarboxylic acid cycle, fueling GIST progression and secondary resistance. Inversely, CCN2 secreted by GIST cells promoted TGFβ1 production in CAFs. These findings uncover a TGFβ1/CCN2/Rack1/PGK1 mechanism linking CAF-mediated metabolic reprogramming to IM resistance in GISTs. Targeting CAF–GIST interactions and key metabolic pathways presents a promising therapeutic strategy. Significance: Cross-talk between cancer-associated fibroblasts and cancer cells promotes metabolic and tumor microenvironmental reprogramming that drives imatinib resistance in gastrointestinal stromal tumors, providing potential targets to overcome resistance and improve therapeutic outcomes.
Background: Long non-coding RNAs (lncRNAs) participate in a wide range of physiological processes, and their dysregulation is prevalent in human cancers, indicating critical roles in tumorigenesis. In intermediate- to high-risk gastrointestinal stromal tumors (GISTs), resistance to tyrosine kinase inhibitors (TKIs) remains a major therapeutic challenge. Therefore, identifying lncRNAs as potential novel therapeutic targets is of considerable interest. Methods: Three paired samples of intermediate- to high-risk GIST tissues and adjacent normal tissues were subjected to transcriptome sequencing. High-content screening (HCS) was subsequently performed to identify candidate lncRNAs with significant effects on GIST cell proliferation. Loss-of-function experiments were conducted, and cell proliferation, migration, and apoptosis were evaluated using the Cell Counting Kit-8 (CCK-8) assay, colony formation assay, Transwell migration assay, and flow cytometry, respectively. In addition, in situ hybridization (ISH) was performed on 507 primary GIST tissue specimens to examine the association between CTD-2245E15.3 expression and clinicopathological features, including progression-free survival (PFS) and overall survival (OS). Results: Transcriptome sequencing revealed 2924 upregulated and 2629 downregulated lncRNAs in GIST tissues compared with adjacent normal tissues. Based on HCS results, CTD-2245E15.3 was selected for further functional analyses. CCK-8 assays demonstrated that knockdown of CTD-2245E15.3 significantly inhibited proliferation of GIST cells. Consistently, colony formation and migratory capacity were markedly reduced in the shCTD-2245E15.3 group compared with controls. Furthermore, flow cytometric analysis showed a significant increase in apoptosis following CTD-2245E15.3 silencing. ISH analysis revealed that high CTD-2245E15.3 expression correlated with adverse clinicopathological features and poorer PFS and OS. Conclusions: Our study demonstrates that CTD-2245E15.3 promotes proliferation and migration of GIST cells and is associated with poor prognosis, highlighting its potential as a therapeutic target and diagnostic biomarker.