591 Background: RAS proteins (such as KRAS, NRAS, HRAS) are small GTPases that drive cell proliferation and survival when bound to GTP. Mutant RAS proteins exist predominantly in the GTP-bound (RAS(ON)) state, leading to excessive downstream signaling via interaction with effectors such as RAF kinases. Oncogenic KRAS is required for the initiation, progression, and maintenance of pancreatic ductal adenocarcinoma (PDAC) (Hezel et al, 2006, Ying et al 2012). Although extinction of KRAS expression as well as pharmacological inhibition of RAS effectors clearly abrogate the growth of human PDAC models, clinical trials of drugs targeting key components of the RAS pathway have remained largely unsuccessful. Several factors contribute to these failures including redundancy in signaling surrogates downstream of KRAS and/or tumor complexity driven by co-occurring genomic alterations and intra-tumoral heterogeneity. Methods: RMC-6236 is a small molecule that binds to an intracellular chaperone protein, Cyclophilin A (CypA), resulting in an inhibitory binary complex that binds active, GTP-bound RAS to form a tri-complex and suppresses RAS signaling by disrupting interactions with effectors such as RAF kinases. Results: Here, we demonstrate that single agent RMC-6236, a first-in-class, orally bioavailable, RAS-selective tri-complex inhibitor of multiple RAS mutations and wild-type RAS (RASMULTI inhibitor) is highly efficacious in preclinical models of KRAS mutant PDAC (with marked activity in RAS-mutant colorectal cancer models described in Koltun et al, AACR 2021). RMC-6236 suppresses phosphorylation of ERK kinases, downstream effectors of RAS involved in cell proliferation, and induces growth suppression and apoptosis in multiple human cancer cell lines in vitro. Oral administration of RMC-6236 produces deep, durable, and dose-dependent suppression of tumor RAS pathway activation in vivo. An extended duration of tumor pharmacodynamic activity, relative to plasma exposure, is observed that likely reflects retention of RMC-6236 in tumor tissue due to high affinity binding to CypA. Daily dosing of RMC-6236 drives profound and durable tumor regressions in multiple cell line derived (CDX) and patient derived (PDX) xenograft models of KRAS mutant PDAC at doses that are well-tolerated. Conclusions: These results indicate that direct targeting of mutant and possibly wild-type RAS in PDAC, without inhibition of signaling nodes outside the canonical RAS pathway, has the potential to translate into clinical benefit for patients with pancreatic cancer harboring mutations in KRAS that may be superior to therapies aimed at upstream or downstream signaling elements within the RAS pathway. Our preclinical data strongly support the inclusion of PDAC patients in our planned clinical trial of RMC-6236 in patients with advanced solid tumors.