ARID1A Deficiency in Diffuse-Type Gastric Cancer Promotes a Pyrimidine Metabolic Vulnerability

Loss-of-function mutations in ARID1A define an aggressive subtype of diffuse gastric cancer (DGC) that is often resistant to standard chemotherapy. Here, we uncover a precise metabolic vulnerability in ARID1A-deficient DGC driven by a specific transporter defect. Through integrated metabolomic and transcriptomic analyses, we demonstrate that ARID1A loss transcriptionally represses the high-affinity nucleoside transporter SLC28A3. Our profiling revealed a critical lack of redundancy in the concentrative transporter family (CNT) in DGC, establishing a strict reliance on SLC28A3 for maintaining intracellular deoxycytidine (dC) pools. Consequently, ARID1A deficiency creates a severe "low-dCTP" metabolic bottleneck. We show that the dC analog Gemcitabine exploits this state through a distinct functional dichotomy: it enters via intact equilibrative transporters (ENTs) to target cells that have lost their competitive dC barrier. Mechanistically, Gemcitabine exerts a "dual-hit" effect by outcompeting the scarce dC pool for DNA incorporation while simultaneously inhibiting ribonucleotide reductase, thereby blocking de novo nucleotide synthesis. This synergistic collapse of pyrimidine metabolism was validated in patient-derived ex vivo cultures and in vivo peritoneal dissemination models. Our findings provide a robust mechanistic basis for repurposing Gemcitabine as a precision therapy for ARID1A-deficient DGC, offering a potent strategy for this intractable malignancy. Overall design: RNA-seq profiling of the non-treatment condition of DGC cell lines. ATAC-seq profiling of DGC cell line treated with siRNA for non-targeting or ARID1A for 72 h.