Abstract Cancer-associated fibroblasts (CAFs) are cells responsible for deposition and remodeling of the extracellular matrix, and modification of the stromal microenvironment, which facilitates tumor growth. Moreover, CAFs are extremely immunosuppressive because they produce large amounts of inhibitory cytokines, growth factors and metabolites. Thus, understanding how CAFs impart severe immunosuppression in the tumor microenvironment is critical to make PDAC amenable to immunotherapies. The link between metabolism and immunosuppression in tumors has become clearer in recent years, and we postulate that one such metabolite, gamma-aminobutyric acid (GABA), a key inhibitory neurotransmitter, can act as a major immunosuppressive metabolite in PDAC. Our preliminary data shows that CAFs produce high levels of glutamate (Glu) and that metabolites derived from glutamate, such as GABA, are immunosuppressive to NK cells. Moreover, we found that CAFs could produce GABA, de novo, as determined by heavy isotope tracing. Thus, our goal is to uncover the mechanisms used by CAFs to produce GABA, and to evaluate the immunosuppressive mechanisms behind CAF- derived GABA. GABA production was measured in the conditioned media of CAFs grown in 3D culture, where it was observed that inhibition of glutaminase (GLS) and glutamine synthetase (GS) enzymes was not able to decrease the GABA levels. In addition, CAFs did not express glutamate decarboxylase (GAD), the canonical enzyme responsible for the conversion of Glu to GABA, which led us to explore the expression of a non-canonical GABA synthesis pathways in CAFs. Intriguingly, we found that CAFs derived from patient tumors expressed the enzymes of non-canonical pathway (ODC1, DAO, ALDH1A1 and ALDH9A1), and to a greater degree when compared to normal fibroblasts isolated from donor pancreata (Gift of Life Program). CRISPRi mediated knockdown (KD) of aldehyde dehydrogenase 1 (ALDH1a1) in CAFs decreased GABA production compared to control CAFs. On the other hand, CAFs knocked down for diamine oxidase (DAO), showed increased GABA production, suggesting a compensatory mechanism for GABA production upstream of ALDH1a1. Overall, our preliminary data demonstrate that CAFs produce GABA de novo, via a putative alternative GABA synthesis pathway. Our next steps are to identify the cytokine profile of KDs CAFs and perform functional assays utilizing our 3D co-culturing system, uncovering natural killer cell responses to tumor cells in the presence of CAFs with and without the enzymes necessary for GABA production. Translationally, we are analyzing the tumor interstitial fluid from PDAC patients to uncover metabolites/cytokines (secreted factors) that can be correlated with patient parameters (ex. overall survival) and immune cell activation and functionality. It is our hope that the insight provided by this study can lead to a novel set of prognostic or diagnostic factors to improve clinical outcomes in this devasting disease. Citation Format: Ariana Musa de Aquino, Myree Graves, Farrah Ali Ali, David A. Rangel, Langley Kyra, Julie Clark, David Kwon, Guillaume Cognet, Alex Muir, Débora Barbosa Vendramini Costa, Ralph Francescone. Identification of a non-canonical pathway used by CAF to deliver gamma-aminobutyric acid in pancreatic cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl_2):Abstract nr C042.