Synergistic oncogenesis in human cholangiocytes: IDH1 mutation and the biliary microenvironment to drive early transformation and a targetable metabolic dependency in patient-derived organoids.
A novel, physiologically relevant in vitro model that recapitulates the synergistic oncogenesis driven by IDH1 R132C mutation and microenvironmental stress in early CCA is established and provides a preclinical platform for identifying and testing interceptive strategies aimed at reversing or halting early cholangiocarcinogenesis in high-risk settings.
Abstract
308
Background:
Cholangiocarcinoma (CCA) is a lethal malignancy with limited understanding of its early developmental mechanisms. A significant subset of CCA harbors gain-of-function mutations in
IDH1
(e.g., R132C), which promotes accumulation of the oncometabolite 2-hydroxyglutarate (2-HG). A major barrier in dissecting early events is the lack of models that faithfully recapitulate human physiology. Patient-derived organoids (PDOs) retain the genetic and phenotypic diversity of the original tissue. Base editing now enables the introduction of specific nucleotide mutations into physiologically relevant models to study causal driver events. However, the oncogenic impact of IDH1 mutations on normal and pre-diseased cholangiocytes, and their synergy with factors like bile acids, are poorly defined.
Methods:
Normal and dilated cholangiocyte organoids were derived from clinical specimens. Cytosine base editing installed the IDH1 R132C mutation. Phenotypes (proliferation, invasion, CEA/CA19-9 secretion) and 2-HG levels were assessed, with/without bile acid challenge. Integrated transcriptomics and metabolomics delineated underlying reprogramming.
Results:
Introduction of the
IDH1
R132C mutation into both normal and dilated cholangiocyte organoids induced a pro-oncogenic phenotype, including enhanced proliferation, invasion, and secretion of CEA/CA19-9, concomitant with 2-HG accumulation. Mechanistically, multi-omics analysis revealed that IDH1 R132C, particularly in dilated cholangiocytes, drives a transcriptional program characteristic of early biliary transformation, involving dysregulation of pathways governing cell fate (e.g., Hippo/YAP suppression) and metabolism (e.g., upregulated glutaminolysis). Crucially, bile acids acted as a potent synergistic cofactor, amplifying phenotypes and rewiring the metabolic network to create a dependency on specific biosynthetic pathways. This combination effectively locked organoids into a proliferative, precursor-like state.
Conclusions:
We have established a novel, physiologically relevant in vitro model that recapitulates the synergistic oncogenesis driven by
IDH1
mutation and microenvironmental stress in early CCA. Our study moves beyond phenomenology to identify the critical early molecular and metabolic nodes activated during this transformation. Importantly, by targeting a key identified metabolic vulnerability (e.g., with a glutaminase inhibitor or a mutant IDH1 inhibitor), we were able to significantly attenuate the observed hyperproliferative and invasive phenotypes, and reduce tumor marker secretion. This work provides a preclinical platform for identifying and testing interceptive strategies aimed at reversing or halting early cholangiocarcinogenesis in high-risk settings.
BACKGROUND & AIMS
Fibroblast growth factor receptor 2 gene (FGFR2) rearrangements are among the most common oncogenic drivers in intrahepatic cholangiocarcinoma (iCCA). While FGFR inhibitors are clinically approved, primary and secondary resistance remain major limitations. Preclinical investigation of resistance mechanisms, including cancer cell-extrinsic crosstalk, is hampered by current models that rely on human FGFR2-fusion transgenes in immunodeficient hosts. We therefore aimed to generate an entirely murine FGFR2-fusion driven iCCA (Ff-iCCA) model to study immunomodulatory mechanisms in the tumor microenvironment (TME).
METHODS
A syngeneic cholangiocyte organoid-based iCCA mouse model was engineered via endogenous chromosomal rearrangement of the Fgfr2 gene combined with Trp53 deletion (PFf) and other co-occurring genetic alterations. KrasG12D-mutated lines (PK) served as comparison. TME characterization was performed using 30-plex spatial proteomics on ∼250,000 cells. Bulk RNA-sequencing was conducted on FGFR inhibitor-treated PFf and PK organoids. Pharmacodynamics of FGFR inhibition on Ff-iCCA were assessed by immunostaining and quantitative RT-PCR.
RESULTS
Intrahepatic implantation produced well-differentiated Ff-iCCA with morphologic features resembling human small duct type iCCA. In immunocompetent hosts, additional co-alterations were required for tumor penetrance, with Pten deletion being most robust with 75%. Compared to KRAS-driven iCCA, Ff-iCCA showed a significantly increased infiltration by Ly-6C/G+ neutrophils (19-fold, p=0.004) and CD8+ T cells (8-fold, p<0.001). Transcriptome analysis revealed increased chemokine expression in PFf versus PK organoids, which was not reversed by FGFR inhibition. Ff-iCCA responded to FGFR inhibition with a 6.5-fold reduced proliferation in vivo (p=0.029), without observation of significant TME remodeling.
CONCLUSIONS
Our syngeneic murine Ff-iCCA model recapitulates hallmarks of human FGFR2-fusion iCCA, providing a platform for functional investigation of cancer cell-TME crosstalk in this molecular subtype.
IMPACT AND IMPLICATIONS
This study introduces the first fully syngeneic, endogenously engineered murine model of FGFR2-fusion driven iCCA, overcoming the key limitation of prior models relying on human transgenes in immunodeficient hosts. The model faithfully recapitulates hallmarks of human FGFR2-fusion iCCA, including a small duct type morphology and a distinct neutrophil-enriched tumor immune microenvironment, validating its translational relevance. The finding that an upregulated chemotaxis signature in FGFR2-fusion persists despite FGFR inhibition, alongside upregulation of interferon-stimulated genes upon treatment, points to compensatory immunomodulatory mechanisms that remain to be mechanistically resolved. Overall, this work provides a physiologically relevant platform to interrogate cancer cell-tumor microenvironment crosstalk in FGFR2-fusion driven iCCA.
Nugzar Lekiashvili, Trinh Kieu Dinh, Alexander Olkus et al.· JHEP Reports· 0 citations
Gain-of-function mutations in isocitrate dehydrogenase 1/2 (IDH1/2) produce the oncometabolite 2-hydroxyglutarate (2-HG) and occur in the majority of lower-grade gliomas as well as in a subset of colorectal cancers. We sought to identify conserved metabolic addictions in IDH-mutant gliomas and CRC that could be pharmacologically targeted, with special emphasis on rare glioma variants.
Untargeted metabolomics and 13C-glutamine isotope tracing were performed on a panel of IDH1-mutant oligodendroglioma, astrocytoma, and CRC cell lines, alongside matched wild-type controls. A CRISPR-Cas9 metabolic gene knockout screen was conducted in IDH1-mutant models. Therapeutic vulnerabilities were validated in vitro and in orthotopic xenografts, including a model derived from a rare IDH-mutant primary spinal cord glioma.
Metabolomic profiling revealed a shared dependence on glutamine anaplerosis to sustain the tricarboxylic acid cycle and 2-HG production in both IDH-mutant gliomas and CRC. CRISPR screening identified glutaminase (GLS) and glutamate dehydrogenase (GLUD1) as essential nodes for mutant cell survival. Pharmacological inhibition of GLS with CB-839 induced metabolic crisis and apoptosis selectively in IDH-mutant cells. Strikingly, the combination of the mutant IDH inhibitor ivosidenib and CB-839 produced synergistic growth inhibition and durable 2-HG reduction in intracranial and subcutaneous tumor models, including complete remission in the rare spinal cord glioma xenograft.
IDH-mutant brain and colorectal tumors converge on glutamine metabolism as a critical liability. Dual blockade of mutant IDH and glutaminolysis represents a novel, cross-entity therapeutic regimen and offers a promising treatment avenue for rare, surgically challenging IDH-mutant gliomas.
Lei Li. Targeting shared metabolic vulnerabilities in IDH-mutant gliomas and colorectal cancer: a new therapeutic angle for rare brain tumors [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Breaking Barriers in the Fight against Rare Cancers; 2026 Jul 18-20; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2026;86(14_Suppl):Abstract nr A014.
Cholangiocarcinoma (CCA), is an aggressive malignancy often diagnosed at an advanced, inoperable stage, with 5-year survival rate of 3%. KRAS mutations are frequent driver alterations linked to poorer prognosis in this cancer. The KRAS G12C mutation inhibitor Adagrasib has been approved by the FDA for the treatment of advanced non-small-cell lung cancer, indicating that the KRAS mutation is a druggable target. However, KRAS mutant inhibitors have yet to be employed in CCA treatment.
A mouse CCA model, designated AKP, was generated by crossing genetically engineered mouse strains harboring Alb-Cre, LSL-KrasG12D, and p53L/L. Subsequently, the AKP-M4 cell line was derived by serially passaging tumor cells from AKP mice across three generations. Sanger sequencing and immunohistochemical (IHC) staining were used to determine the KRAS G12D mutation and cytokeratin 19, respectively. MTT assay was used to assess the cell proliferation inhibition by MRTX-1133. Immune cell profiles in AKP-M4-bearing mice treated with MRTX-1133 were determined by flow cytometry.
Sanger sequencing confirmed G12D mutation in the KRAS gene in AKP-M4 cells. IHC analysis demonstrated that AKP-M4 tumors expressed cytokeratin 19, a key marker for CCA. The in vitro proliferation of AKP-M4 cells, but not KRAS wild-type SNU-1079 and SSP-25 cells, was reduced by the KRAS G12D mutation inhibitor MRTX-1133. We further assessed the anti-cancer efficacy of MRTX-1133 in AKP-M4-bearing C57BL/6 mice, observing a significant dose-dependent reduction in tumor volume and weight. Flow cytometry analysis showed that MRTX-1133 elevated CD4+ T cells, CD8+ T cells, and M1 macrophages, while decreasing MDSCs and M2 macrophages. Notably, PD-L1 expression on tumor cells was diminished.
KRAS G12D mutant inhibition reduces cancer cell proliferation and stimulates immune cells. This supports the development of KRAS mutation inhibitor for CCA treatment.
National Science and Technology Council, Taiwan
Tumor Immunology: Cellular Responses and Tumor Microevironment (TIME)
J. Hung, Tsai-Hsien Hung, Chun-Nan Yeh et al.· Journal of Immunology· 0 citations
Triple-negative breast cancer (TNBC) is an aggressive subtype with limited therapeutic options and poor prognosis. The early pre-malignant events that lead to TNBC remain elusive, largely due to a lack of tractable models that recapitulate stepwise tumor initiation in a natural tissue context. Here, we use a spontaneous mouse model with mammary-specific deletion of Brca1 and Trp53, which develops TNBC, together with single cell RNA sequencing and multi-color flow cytometry to investigate the cellular and molecular landscape of the pre-tumoral mammary glands. Strikingly, the microenvironmental changes occur before any detectable tumors, driven by Brca1/Trp53 loss and MYC activation in luminal epithelial cells. Although macrophages remained proportionally comparable in the pre-tumoral lesions, their interactions with other populations were much reinforced. Brca1/Trp53 loss generated unstable genome and DNA fragment due to impaired DNA damage repair, which was sensed by macrophages and triggered type I interferon expression. The reprogrammed macrophages expressed IL-10 to paralyze cytotoxic T cells, TGFβ to target mast cells, and CD137L to stimulate Tregs, contributing to a favorable tumor-initiating niche. The rewired macrophages were stimulated by mutant luminal progenitor-derived CSF1 and were continuously recruited via CCR1-CCL3/5 axis. Such remodeling of mammary gland milieu allows breast tumor formation. After tumor was established, neutrophils were expanded and recruited to the developed tumors, circulation, and the lungs regardless of metastasis outgrowth. The neutrophils were amplified by elevated CSF3 from abnormal luminal progenitors and SAA2 from secretory alveolar cells in early lesions, which created chronic inflammation and facilitated the establishment of pre-metastatic niche in the lungs. Together, our study establishes a stepwise model of immune remodeling preceding TNBC onset and uncovers macrophage-centered signaling as a mechanistic axis linking genetic instability to a tumor-permissive niche. These findings provide insights into early intervention targets in high-risk breast cancer.
LUQI HUANG, James A. Miller, Hanisah MK, Bernett Lee, Su I-hsin. Early Immunosuppressive Landscape Shaped by BRCA1/p53 Loss in Early Pre-Malignant Mammary Glands Permits Tumor Development [abstract]. In: Proceedings of Frontiers in Cancer Science 2025; 2025 Nov 5-7; Singapore. Philadelphia (PA): AACR; Cancer Res 2026;86(13_Suppl):Abstract nr P10.
Luqi Huang, J. A. Miller, Hanisah Mk et al.· Cancer Research· 0 citations
Pancreatic Ductal Adenocarcinoma (PDAC) has a dismal prognosis and limited treatment options. With ∼90% of PDAC cases harboring KRAS mutations, targeting KRAS has been a major therapeutic focus. However, its small size and high affinity for GTP present challenges for direct inhibition. Our lab previously identified Merodantoin (C1), a compound that paradoxically hyperactivates mutant KRAS and induces apoptosis, particularly in colon cancer. This study explores the efficacy and mechanism of C1 in KRAS-mutant PDAC. We treated PDAC cell lines and patient-derived organoids (PDOs) with C1 and assessed cell viability, ROS production, iron accumulation, and protein expression via flow cytometry and Western blotting. C1 exhibited selective cytotoxicity toward PDAC cells harboring KRAS G12C/G12D mutations, sparing wild-type KRAS cells. Mechanistically, C1 elevated both total and mitochondrial ROS and labile Fe2+ in all cells, but only induced lipid peroxidation and ferroptosis in KRAS G12-mutant cells. Ferroptosis inhibitors (ferrostatin-1, liproxstatin) and iron chelator (DFO) rescued these effects. C1 suppressed the GSH-antioxidant system selectively in mutant KRAS cells by reducing cystine uptake, GPX4 expression, and GSH levels in MIA PaCa-2 and PANC-1, but not in wild-type BxPC-3 cells. Resistance in BxPC-3 was linked to higher basal GPX4 levels. Silencing GPX4 sensitized BxPC-3 cells to C1, confirming its role in resistance. Furthermore, KRAS knockdown or inhibition (e.g., sotorasib) abrogated C1-induced ferroptosis, confirming KRAS as an upstream driver of this death pathway. These findings were validated in PDOs, with ongoing work in mouse models. Overall, this study presents C1 as a novel therapeutic agent for KRAS-mutant PDAC and introduces the concept of hyperactivating oncogenic KRAS to selectively induce ferroptotic cancer cell death.
Haiyuxin Zhu, Kartini Iskandar, Nur Syafiqah Binte Sulaiman, Benedict Joseph, Anne-sophie Armand, Franck Oury, Shazib Pervaiz. Drug-induced Hyperactivation of the oncogene KRAS induces ferroptosis in G12-mutant KRAS Pancreatic Ductal Adenocarcinoma (PDAC) [abstract]. In: Proceedings of Frontiers in Cancer Science 2025; 2025 Nov 5-7; Singapore. Philadelphia (PA): AACR; Cancer Res 2026;86(13_Suppl):Abstract nr P87.
Haiyuxin Zhu, K. Iskandar, Nur Syafiqah Binte Sulaiman et al.· Cancer Research· 0 citations