Objectives/Goals: To investigate how distinct KRAS mutations (G12D, G12V, and G12R) drive distinct molecular and phenotypic programs in pancreatic tumorigenesis, linking early lineage and signaling differences to clinical patterns such as stage at diagnosis, nodal status, and survival to ultimately inform allele-specific therapies. Methods/Study Population: We analyzed clinical cohorts (MSK-IMPACT n=1,360; COMPASS n=100) and conducted lineage-tracing studies in mice carrying KRAS-G12D, KRAS-G12V, or KRAS-G12R alleles under inflammatory, genetic, and pharmacologic perturbations. Spatial transcriptomic and proteogenomic profiling of resected PDAC tissues characterized EMT and immune signaling states. Functional studies assessed EGFR–PI3K/AKT–RAC1/VAV1 signaling, including RAC1 inhibition and constitutive AKT activation, to reveal allele-specific susceptibilities. Results/Anticipated Results: KRAS-G12D drove robust lineage plasticity, enhancer remodeling, and early neoplastic progression. KRAS-G12R/V showed limited EMT and increased inflammatory signatures, with impaired EGFR–RAC1/VAV1 activation restricting lineage reversion. Constitutive AKT activation rescued the tumor-initiating defect of KRAS-G12R in vivo, pinpointing a downstream signaling bottleneck, which could be exploited therapeutically. Spatial and transcriptomic profiling of resected human PDAC revealed that KRAS-G12R tumors were more often early-stage, node-negative, and linked to improved survival relative to KRAS-G12D. Discussion/Significance of Impact: Integrating patient and mouse data reveals a mechanistic hierarchy among KRAS alleles that governs lineage fate, signaling reliance, and clinical course, pointing to opportunities for allele-tailored therapeutic strategies in pancreatic cancer.

Inflammation in the pancreas drives acinar-to-ductal metaplasia (ADM), a progenitor-like state that can be hijacked by mutant Kras in the formation of pancreatic ductal adenocarcinoma. How these cell fate decisions vary according to KRAS mutation remains poorly understood. To define mutation-specific lineage reversion and tumor initiation, we implement Ptf1a-tdTomato mice and multiple KRAS mutants across several genetic, pharmacologic, and inflammatory perturbations in vivo. Whereas KRASG12D co-opts injury to enable lineage reversion, enhancer reprogramming, and tumor initiation, KRASG12R/V cannot sustain dedifferentiated and neoplastic transcriptional and epigenetic programs. Specifically, KRASG12R/V mutants fail to invoke robust EGFR, AKT, and RAC1/VAV1 signaling and to license Pou2f3 and Vav1 in chromatin, such that only constitutive AKT activation is sufficient to rescue the tumorigenic potential of KRASG12Rin vivo. As the marked heterogeneity among KRAS variants begins early in tumorigenesis, these data are crucial to deciphering mutation-specific oncogenic trajectories and directing the implementation of KRAS-directed therapeutics.

Inflammation is essential to the disruption of tissue homeostasis and can destabilize the identity of lineage-committed epithelial cells. Here, we employ lineage-traced mouse models, single-cell transcriptomic and chromatin analyses, and CUT&TAG to identify an epigenetic memory of inflammatory injury in the pancreatic acinar cell compartment. Despite resolution of pancreatitis, our data show that acinar cells fail to return to their molecular baseline, with retention of elevated chromatin accessibility and H3K4me1 at metaplasia genes, such that memory represents an incomplete cell fate decision. In vivo, we find this epigenetic memory controls lineage plasticity, with diminished metaplasia in response to a second insult but increased tumorigenesis with an oncogenic Kras mutation. The lowered threshold for oncogenic transformation, in turn, can be restored by blockade of MAPK signaling. Together, we define the chromatin dynamics, molecular encoding, and recall of a prolonged epigenetic memory of inflammatory injury that impacts future responses but remains reversible.

Tissue homeostasis depends on responses to environmental insults to restore cellular phenotype, microenvironment composition, and tissue architecture. Inflammation is essential to the disruption of homeostasis, and, in the pancreas, can destabilize the identity of terminally differentiated acinar cells. Herein we employ lineage-traced mouse models to delineate the chromatin dynamics that accompany the cycle of metaplasia and regeneration following pancreatitis, and unveil the presence of an epigenetic memory of inflammation in the pancreatic acinar cell compartment. We observe that despite histologic resolution of pancreatitis, acinar cells fail to return to their molecular baseline after several months, representing an incomplete cell fate decision. In vivo, this epigenetic memory controls lineage plasticity, with diminished metaplasia in response to a second inflammatory insult but increased tumorigenesis with an oncogenic Kras mutation. We demonstrate that both persistent chromatin and transcriptional changes constituting memory are recalled with oncogenic stress. Together, our findings define a capacity for an environmental insult to control future cell-fate decisions in a context-dependent manner. The ability of epigenetic memory to potentiate tumor initiation both broadens the relationship between inflammation and cancer and raises the possibility that inducing epigenetic ‘amnesia’ of an inflammatory insult could be leveraged as a novel cancer prevention strategy. Citation Format: David J. Falvo, Adrien Grimont, Paul Zumbo, Julie L. Yang, Alexa Osterhoudt, Grace Pan, Andre F. Rendeiro, John Erby Wilkinson, Friederike Dundar, Olivier Elemento, Rhonda K. Yantiss, Doron Betel, Richard Koche, Rohit Chandwani. An epigenetic memory of inflammation controls context-dependent lineage plasticity and KRAS-driven tumorigenesis in the pancreas. [abstract]. In: Proceedings of the AACR Special Conference: Cancer Epigenomics; 2022 Oct 6-8; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2022;82(23 Suppl_2):Abstract nr PR008.

SUMMARY Inflammation is essential to the disruption of tissue homeostasis, and, in the pancreas, can destabilize the identity of terminally differentiated acinar cells. Herein we employ lineage-traced mouse models to delineate the chromatin dynamics that accompany the cycle of metaplasia and regeneration following pancreatitis, and unveil the presence of an epigenetic memory of inflammation in the pancreatic acinar cell compartment. We observe that despite histologic resolution of pancreatitis, acinar cells fail to return to their molecular baseline after several months, representing an incomplete cell fate decision. In vivo , this epigenetic memory controls lineage plasticity, with diminished metaplasia in response to a second inflammatory insult but increased tumorigenesis with an oncogenic Kras mutation. We demonstrate that both persistent chromatin and transcriptional changes constituting memory are recalled with oncogenic stress. Together, our findings define the dynamics and recall of an epigenetic memory of inflammation that impacts cell fate decisions in a context-dependent manner.

The most prominent KRAS variants (G12D, G12V, G12R) that together represent around 80% of patients with pancreatic ductal adenocarcinoma (PDAC) are so far undruggable. While the role of several Kras mediators have been characterized in the initiation of PDAC, very little is known about the hierarchy of Kras effectors in the maintenance of the tumor. In recent years, strategies targeting the effectors downstream of mutant KRAS have offered scope for combined inhibition of EGFR and CRAF or MEK/RAF. However, these studies, unfortunately, have been limited by either narrow interrogation of downstream effectors in mouse models, or by the use of two-dimensional cell culture systems that may not recapitulate dependencies of the tumor in situ, respectively. To systematically interrogate the potential molecular dependencies in pancreatic tumor maintenance across several combinations of driver mutations, we have deployed in vitro and in vivo approaches in which we have selectively targeted key mediators of known KRAS-dependent pathways. These include Craf, Braf, PI3K, RalA, RalB, and Rac1, for which we have used inducible GFP-coupled shRNAs in 3D mouse and human tumor organoids harboring KrasG12D and p53R172H mutation. Using competition, cell cycle, and volumetric assays, we have uncovered that Rac1, Kras, and Craf are essential to the growth of PDAC organoids, whereas Braf, PI3K, RalA and RalB are dispensable. Interestingly, Rac1 depletion led to the strongest phenotype among the Kras mediators with a reduction of macropinocytosis, cell migration and colony formation in vitro. In an orthotopic pancreatic injection model, we observed that Rac1 inhibition in vivo led to diminished primary tumor growth, improved survival, and a reduction of metastatic incidence and outgrowth. In parallel, we performed RNA-sequencing on Kras-, Craf-, Rac1-, RalA- and Renilla-depleted organoids and found that Rac1 depletion rewires tumor cells to acquire a more PanIN-like phenotype, highlighting the importance of these proteins for the maintenance of PDAC cells. We also identify in Rac1- depleted cells evidence of deregulation of reactive oxygen species (ROS) and induction of a senescence-associated secretory phenotype (SASP) compared to control organoids. With a cytokine and chemokine array, we confirmed the increase of SASP chemokines (Csf3, Cxcl1, Cxcl2 and Cxcl5) and also detected bona fide senescence via SA-βgal staining. Finally, using several Rac1 inhibitors, we recapitulate the importance of Rac1 in PDAC growth. These data suggest that among the pleiotropic signaling downstream of mutant Kras, Rac1 is a critical node in PDAC maintenance that promotes tumor cell proliferation and senescence escape. Our findings point towards future efforts to couple Rac1 inhibition to define therapeutic synergies with immunotherapy and/or radiation. All together, we anticipate these findings can inform the subsequent development of novel therapies to address these vulnerabilities. Citation Format: Adrien Grimont, David J. Falvo, Paul Zumbo, Grace Pan, John Nguyen, Rhonda K. Yantiss, Doron Betel, Laura Martin, Steven D. Leach, Rohit Chandwani. Rac1 is essential for the maintenance of established KrasG12D-driven pancreatic ductal adenocarcinoma through senescence escape [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr B055.

Inflammation is essential to the disruption of tissue homeostasis, and, in the pancreas, can destabilize the identity of terminally differentiated acinar cells. A longstanding question has been whether a history of past injuries instructs subsequent homeostatic responses to future stimuli – despite being distantly separated in time. Thus, we employ here Mist1-CreERT2; LSL-tdTomato lineage-tracing mice to investigate the long-term effects of a transient inflammatory episode on pancreatic tissue homeostasis. We delineate the chromatin dynamics that accompany the cycle of metaplasia and regeneration following pancreatitis, and reveal that the pancreatic acinar cell compartment durably retains specific inflammation-induced epigenetic changes even 18-weeks after exposure to the original inflammatory stimulus. We observe that despite histologic resolution of pancreatitis, acinar cells fail to return to their molecular baseline after several months, representing an incomplete cell fate decision. Motif analysis demonstrates the enrichment of AP-1/Fra1 motifs at these persistently accessible memory regions—a transcriptional effector activated downstream of the Ras/MAPK pathway. This epigenetic memory controls lineage plasticity, with diminished metaplasia in response to a second inflammatory insult but increased tumorigenesis with an oncogenic Kras mutation. We demonstrate that pancreatic acinar cells exhibit rapid malignant transformation upon re-challenge with oncogenic stress via inflammatory memory recall, with robust reactivation of genes associated with differentially accessible memory regions. Together, our findings define the dynamics and recall of an epigenetic memory of inflammation that impacts cell fate decisions in a context-dependent manner. Citation Format: David J. Falvo, Adrien Grimont, Paul Zumbo, Julie L. Yang, Alexa Osterhoudt, Grace Pan, Andre F. Rendeiro, John E. Wilkinson, Friederike Dündar, Olivier Elemento, Rhonda K. Yantiss, Doron Betel, Richard Koch, Rohit Chandwani. An epigenetic memory of inflammation controls context-dependent lineage plasticity in the pancreas [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer; 2022 Sep 13-16; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2022;82(22 Suppl):Abstract nr PR009.

Tissue homeostasis depends on responses to environmental insults to restore cellular phenotype, microenvironment composition, and tissue architecture. Inflammation is essential to the disruption of homeostasis in epithelial tumorigenesis, but how a temporally remote inflammatory episode impacts tumor development is unknown. Herein we employ lineage-traced mouse models to unveil the presence of an epigenetic memory of inflammation. We observe that despite histologic resolution of pancreatitis, acinar cells fail to return to the same molecular baseline. In vivo, the memory is associated with diminished metaplasia in response to a second inflammatory insult but increased tumorigenesis when instead subjected to an oncogenic Kras mutation. We find that memory is a cell-intrinsic property, primarily encoded in chromatin, that features persistent derepression of metaplastic genes and is recalled with oncogenic stress. Together, our findings define a capacity for an environmental insult to potentiate future tumor initiation, broadening the relationship between inflammation and cancer.