ABSTRACT
We report a multicystic intrahepatic neoplasm in a 54‐year‐old Japanese woman, representing a previously unrecognized subtype. Grossly, the lesion was a well‐demarcated multicystic tumor with focal papillary projections. Histologically, the cysts were lined by columnar to cuboidal neoplastic cells with brush border‐like luminal microvilli, abundant granular eosinophilic cytoplasm, and small round nuclei. The cystic lumina contained colloid‐like secretion, and bile‐filled glands were occasionally observed. The septa were composed of thin hepatocellular parenchyma. Immunohistochemically, the tumor cells were positive for CD10, CK7, CK19, EpCAM, and SPINK1 and negative for HepPar1, MUC1, MUC2, MUC5AC, and MUC6. Whole‐exome sequencing identified a pathogenic somatic KRAS p.G12V variant. RNA sequencing detected no PRKACA/B fusions. Single‐cell spatial transcriptomics demonstrated that tumor cells clustered most closely with septal and medium‐sized interlobular bile ducts. Gene set activity analysis showed significant suppression of gene sets downregulated by KRAS activation and upregulation of KRAS dependency signature gene sets in tumor cells. These findings distinguish this lesion from established entities of intrahepatic biliary cystic neoplasms, including intraductal papillary neoplasm, intraductal tubulopapillary neoplasm, intraductal oncocytic papillary neoplasm, and mucinous cystic neoplasm. We propose the designation “eosinophilic biliary cystic neoplasm of the liver” for this distinct intrahepatic biliary neoplasm.
Keywords: eosinophilic cystic neoplasm, interlobular bile duct, intrahepatic biliary neoplasm, KRAS pathogenic variant
A distinct multicystic intrahepatic neoplasm composed of low‐grade tumor cells with brush border‐like microvilli and abundant granular eosinophilic cytoplasm is described, for which we propose the term “eosinophilic biliary cystic neoplasm of the liver.”
The tumor harbors KRAS p.G12V driver mutation.
Immunohistochemical and spatial transcriptomic analyses support differentiation of the tumor toward septal and medium‐sized interlobular bile ducts.
Abbreviations
- CT
computed tomography
- GSEA
Gene set enrichment analysis
- IOPN
intraductal oncocytic papillary neoplasm
- IPNB
intraductal papillary neoplasm of the bile ducts
- ITPN
intraductal tubulopapillary neoplasm
- MCN
mucinous cystic neoplasm
- MRI
magnetic resonance imaging
- WES
whole‐exome sequencing
1. Introduction
Established biliary cystic neoplasms of the liver include intraductal papillary neoplasm of the bile ducts (IPNB), intraductal tubulopapillary neoplasm (ITPN), intraductal oncocytic papillary neoplasm (IOPN), and mucinous cystic neoplasm (MCN) [1, 2, 3, 4] Type 1 IPNB is classified by epithelial differentiation into intestinal, pancreaticobiliary, and gastric types [1, 4]. IOPN is characterized by PRKACA/B fusions [5, 6]. MCN is defined by the presence of ovarian‐type subepithelial stroma [2, 3]. IOPN and MCN typically exhibit a gastric phenotype [1, 2, 3].
Herein, we describe an intrahepatic biliary neoplasm composed of eosinophilic cells, showing cystic and focal papillary growth and likely arising from the intrahepatic bile ducts. Because this lesion does not fit any of the established intrahepatic cystic biliary neoplasms, we propose the term “eosinophilic biliary cystic neoplasm of the liver” for this previously unrecognized entity.
2. Clinical Summary
A 54‐year‐old Japanese woman was admitted for evaluation of fever and a urinary tract infection, and a CT incidentally revealed an approximately 12‐cm multicystic hepatic lesion. Laboratory tests showed elevated inflammatory markers and mild cholestatic abnormalities: white blood cell count, 13,100/µL; C‐reactive protein, 12.15 mg/dL; aspartate aminotransferase, 39 U/L; alanine aminotransferase, 44 U/L; γ‐glutamyl transpeptidase, 71 U/L; alkaline phosphatase, 137 U/L; total bilirubin, 1.7 mg/dL; and direct bilirubin, 0.5 mg/dL.
Dynamic contrast‐enhanced CT demonstrated an 11.3‐cm multilocular cystic lesion with a solid component that exhibited early enhancement (Figure 1A). The cystic contents showed high to markedly high signal intensity on T2‐weighted imaging and low to high signal intensity on T1‐weighted imaging, manifesting a “stained‐glass appearance”. The solid component showed mildly high signal intensity on fat‐suppressed T2‐weighted imaging (Figure 1B), with no evidence of restricted diffusion on magnetic resonance imaging (MRI). Based on these imaging findings, hepatic MCN was considered the primary differential diagnosis. Retrospective review of an MRI performed 8 years earlier revealed that the corresponding lesion measured 1.4 cm and presented as a cluster of small cysts (Figure 1C). No communication between the cystic lesion and the bile duct was demonstrated on any imaging modality.
Figure 1.
Imaging findings. (A) Late arterial phase of contrast‐enhanced CT reveals an 11.3‐cm multilocular cystic lesion exophytically protruding from segment 5 of the liver, with early enhancement in the solid component (yellow arrow). (B) On the fat‐suppressed T2‐weighted image, the cystic contents exhibit high to markedly high signal intensity, presenting a “stained‐glass appearance”. The solid component shows mild hyperintensity (yellow arrow). (C) Magnetic resonance cholangiopancreatography with maximum intensity projection performed 8 years ago shows a 1.4‐cm lesion appearing as a cluster of small cysts (yellow arrow).
After percutaneous transhepatic portal vein embolization, the patient underwent right hepatectomy. She had an uneventful postoperative course and remains recurrence‐free at 10 months.
The patient was obese (body mass index 44.3 kg/m2), with a smoking history of five cigarettes/day for 5 years and social alcohol use. Her medical history included hepatic steatosis, urolithiasis, hypertension, diabetes mellitus, gastroesophageal reflux disease, asthma, and cholecystectomy 25 years earlier. Family history was notable for alcoholic liver cirrhosis in her father and pancreatic cancer in her sister.
3. Pathological Findings
3.1. Macroscopic Findings
A well‐demarcated 12‐cm multicystic tumor with exophytic growth was located in segment 5 (Figure 2A). Most cyst walls were flat, with focal whitish papillary growth (Figure 2B). Some cystic spaces contained brown, colloid‐like material. No communication was identified between the multicystic tumor and the bile ducts. There was no bile duct dilatation.
Figure 2.
Macroscopic findings. (A) Gross examination revealed a well‐demarcated, lobulated cystic tumor measuring 12 cm in greatest diameter, located in hepatic segment 5 (yellow arrows). (B) The cut surface showed a multicystic architecture. Most cyst walls were flat; however, focal papillary or solid growth was observed in a subset of cysts (yellow square).
3.2. Microscopic Findings
The cysts were lined by columnar to cuboidal epithelial cells showing predominantly flat or low papillary growth, with occasional branching papillary structures (Figure 3A–D). Complex, arborizing papillary or villous architectures were absent. The neoplastic cells had abundant granular eosinophilic cytoplasm with small, round and uniform nuclei (Figure 3E). The nuclei were centrally to apically located rather than basally positioned (Figure 3C). The apical surface exhibited brush border‐like short luminal microvilli.
Figure 3.
Microscopic findings. (A, B) The cysts were lined by uniform columnar to cuboidal neoplastic cells, showing predominantly flat or low papillary growth, and the cyst wall was composed of hepatocellular parenchyma. (B) is a high‐power view of the boxed area in (A) (×200). (C) Focal papillary structures were observed (microscopic image of the yellow‐boxed area in Figure 2B). (D) In the papillary areas (the black‐boxed area in (C)), a proliferation of uniform eosinophilic tumor cells was observed along delicate thin fibrovascular cores (×200). (E) The tumor cells had brush border–like short luminal microvilli and showed abundant granular eosinophilic cytoplasm with small, round nuclei (×400). (F) The cystic septa were composed of thin hepatic parenchyma (×50).
The cystic lumina contained eosinophilic colloid‐like secretion with vacuoles, and bile‐filled glands were occasionally observed. The cystic septa consisted of thin hepatic parenchyma and showed a paucity of pre‐existing portal tracts, suggesting cystic expansion by replacement of interlobular bile ducts within portal tracts (Figure 3F). At the interface between the hepatic parenchyma and the eosinophilic tumor cells, a thin layer of collagen fibers was partly present, and no infiltrative growth was identified. No ovarian‐type stroma was observed. The background liver parenchyma showed very mild perivenular macrovesicular steatosis.
3.3. Immunohistochemical Findings
To further characterize the eosinophilic tumor cells, immunohistochemical analysis using antibodies listed in Table S1, along with PAS and Alcian blue staining, was performed.
The apical surface of the tumor cells was positive for PAS and Alcian blue, without intracytoplasmic mucin, resembling interlobular cholangiocytes (Figure 4A).
Figure 4.
Results of special stain and immunostains. (A) The apical surface of the tumor cells was positive for Periodic acid–Schiff without intracytoplasmic staining (×200). (B–F, ×200) Tumor cells were positive for CK7 (B), CK19 (C), CD10 (D), EpCAM (E), and SOX9 (F).
The tumor cells were positive for CD10 (highlighting surface microvilli), CK7, CK19, EpCAM, SOX9, and SPINK1, and negative for CD56, CDX2, GATA3, HepPar1, MUC1, MUC2, MUC5AC, MUC6, PAX8 (Figure 4B–F and Table S2). Overall, this immunoprofile more closely resembled that of septal bile ducts (100–300 μm in diameter) and medium‐sized interlobular bile ducts (40–100 μm) than that of small‐sized interlobular bile ducts (15–40 μm), bile ductules (< 15 μm), or large bile ducts (> 300 μm) (Table S2).
3.4. Genome Sequencing Findings
To characterize genomic alterations, whole‐exome sequencing (WES) and RNA sequencing were performed: somatic variants were identified using MuTect2 (GATK, Broad Institute) and annotated with ANNOVAR (Data S7).
WES identified several somatic variants, including KRAS p.G12V (variant allele frequency [VAF], 0.215), ASXL1 p.P752A and p.P813A (0.262), AFF4 p.A570G (0.24), VWF p.R2379H (0.18), FKBP9 p.C102X and p.C155X (0.18), UNC5CL p.P409S (0.144), and PSG5 p.R222X (0.127). Among these, KRAS p.G12V was classified as a pathogenic oncogenic driver variant based on ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/) and OncoKB (https://www.oncokb.org/) annotation.
RNA sequencing revealed no pathogenic rearrangements, including PRKACA/B fusions. Gene set enrichment analysis (GSEA) showed enrichment of a RAS dependency signature (normalized enrichment score 1.616), likely as a result of KRAS activating mutation (Figure S1) [7].
3.5. Single‐Cell Spatial Transcriptome Analysis
Single‐cell spatial transcriptomic analysis was performed on 5‐µm formalin‐fixed, paraffin‐embedded sections using the CosMx platform. Cell segmentation was refined using Proseg v3.0.17 [8], and downstream single‐cell analyses were conducted in R v4.5.1 with Seurat v5.3.1 (https://www.R-project.org/) [9]; details are provided in Data S2.
A total of 38,204 cells were clustered into seven major cell types (Figure S2A–B). Based on spatial distribution, epithelial cell clusters annotated as hepatocytes or cholangiocytes (Clusters 1, 2, 3, 5, and 10) were classified into four groups: tumor, hepatocytes, septal/medium‐sized interlobular bile ducts, and small interlobular bile ducts/bile ductules (Figure S2C–D). Large bile ducts were not included in the analyzed CosMx tissue specimen. Hierarchical clustering demonstrated that tumor cells were most similar to septal/medium‐sized interlobular bile ducts rather than small‐sized interlobular bile ducts/bile ductules, or hepatocytes (Figure 5A). Pathway analysis showed suppression of gene sets downregulated by KRAS activation and upregulation of KRAS dependency signatures in tumor cells (Figure 5B). Among KRAS‐induced genes, ETV4, PIGR, and ST6GAL1 were significantly overexpressed in tumor cells (Figure 5C, Figure S2E). GSEA demonstrated significant enrichment of MYC target pathways, a major downstream effector program of KRAS signaling, in tumor cells compared with septal/medium‐sized interlobular bile ducts (Table S3, adjusted FDR = 0.01). Differential expression results were summarized in Figure 5D and Figure S2F.
Figure 5.
Single‐cell spatial transcriptome analysis. (A) In hierarchical clustering analysis using Pearson correlation as the distance metric and average linkage for both rows (genes) and columns (cell types), tumor cells clustered most closely with septal and medium‐sized interlobular bile ducts. (B) Cell‐level gene set activity analysis using AUCell (v1.30.1) demonstrated significant suppression of hallmark gene sets downregulated by KRAS signaling in tumor cells relative to hepatocytes (p < 0.00001, effect size = −0.70), septal and medium‐sized interlobular bile ducts (p < 0.00001, −0.26), and small‐sized interlobular bile ducts and bile ductules (p < 0.0001, −0.09). Furthermore, KRAS dependency signature gene sets were significantly upregulated in tumor cells compared with hepatocytes (p < 0.00001, 0.33), septal and medium‐sized interlobular bile ducts (p < 0.00001, 0.09), and small‐sized interlobular bile ducts and bile ductules (p < 0.0001, 0.38). (C) Expression of ETV4 in each epithelial type. This KRAS‐induced gene showed significantly increased expression in tumor cells compared with other hepatic epithelial cells (p = 0.0004). (D) Volcano plot of differential expression analysis between tumor cells and septal and medium‐sized interlobular bile ducts. Green square‐marked genes represent MYC pathway‐related genes.
4. Discussion
The current WHO classification recognizes major categories of intrahepatic biliary neoplasms: IPNB, ITPN, IOPN, and MCN [1, 2, 3, 4]. The present lesion does not fit within any of these established entities.
Given the prominent eosinophilic cytoplasm, IOPN is the leading differential diagnosis [10]. However, the tumor lacked expression of MUC5AC, MUC6, and HepPar1 and showed no pathogenic PRKACA or PRKACB alterations, features typically seen in IOPN [5, 6].
Because this lesion shows cystic and papillary growth with continuity to peripheral intralobular bile ducts rather than to large bile ducts, and harbors a KRAS driver alteration, its distinction from the spectrum of type 1 IPNB may be a potential diagnostic issue. Most type 1 IPNBs, as well as their pancreatic counterpart, intraductal papillary mucinous neoplasms, exhibit a gastric phenotype with at least focal MUC5AC expression, and acquisition of this phenotype is considered an early event in tumorigenesis [11]. In contrast, the present case is composed of uniform eosinophilic tumor cells lacking gastric differentiation and MUC5AC expression, supporting its classification as a distinct entity from type 1 IPNB.
This tumor is composed of biliary‐type epithelium expressing CK7, CK19, SOX9, and CD10 with papillary growth. Pancreaticobiliary‐type IPNB should be considered in the differential diagnosis [3]. However, IPNBs typically arise in and communicate with the large bile ducts and lack the abundant granular eosinophilic cytoplasm seen in this lesion. Furthermore, most pancreaticobiliary‐type IPNBs are high‐grade and express MUC1, MUC5AC, and MUC6. Even low‐grade pancreaticobiliary‐type IPNBs express MUC1 [12]. Cystic change in IPNB is associated with mucin hypersecretion, whereas the cyst contents in the present case were colloid‐like [13]. Taken together, these characteristics of pancreaticobiliary‐type IPNB are clearly distinct from those observed in the present case.
In addition, ovarian‐type subepithelial stroma, a defining feature of MCN, was absent [2]. Architecturally, the compact tubular growth pattern and frequent high‐grade cytology typical of ITPN were not observed [3].
Other differential diagnoses include bile duct adenoma and biliary adenofibroma. The cytomorphology of the oncocytic variant of bile duct adenoma is similar to that observed in the present case [3, 14, 15]. However, bile duct adenoma is a small nodular lesion composed of a proliferation of small bile ducts within fibrous stroma and lacks cystic change or bile production. Biliary adenofibroma is defined as a solid–microcystic neoplasm lined by non–mucin‐secreting biliary epithelium and supported by a fibrotic stromal scaffold [16]. The eosinophilic cytoplasm observed in the present case is not a typical feature of biliary adenofibroma. Moreover, the characteristic fibrotic stromal component that is a defining feature of biliary adenofibroma is absent in the present tumor. While biliary adenofibroma has been suggested to show morphological similarity to and an association with von Meyenburg complex, these features are not present in the current case [17, 18]. The pathological features of these entities are distinct from those of the present case.
The tumor showed some morphological similarity to papillary renal neoplasm with reverse polarity, which often harbors KRAS codon 12 alterations [19]. However, the absence of renal tumor history and lack of GATA3 and PAX8 expression exclude metastatic renal origin.
Biliary epithelium is heterogeneous, varying in morphology, phenotype, and function by duct size and location. It is classified into the bile canaliculus, canals of Hering, ductules, interlobular bile ducts, septal bile ducts, and large bile ducts, including area bile ducts (300–400 μm), segmental bile ducts (400–800 μm), and hepatic ducts (> 800 μm) [20, 21, 22]. We found that large duct cells focally expressed MUC2 or MUC6, markers that were completely negative in the tumor cells. Bile ductules expressed CD56, whereas it was negative in larger bile duct and tumor cells. Therefore, tumor cells in the present case lacked gastric‐type mucin expression and showed an immunophenotypic profile similar to that of septal and interlobular bile ducts. Concordantly, spatial transcriptomic profiling demonstrated that tumor cells clustered most closely with septal and medium‐sized interlobular bile ducts. Taken together, these findings support differentiation toward, and possibly origin from, cholangiocytes of intrahepatic bile ducts at the septal/medium interlobular level.
The mild cytological atypia and relatively simple architectural features, together with the low proliferative activity, indicate that this tumor is low grade. The indolent clinical course and the absence of driver alterations other than KRAS mutation are also consistent with low malignant potential. The molecular mechanisms underlying the characteristic morphology, including the eosinophilic cytoplasm, remain unclear and require further investigation.
In conclusion, this case represents a previously unrecognized distinct intrahepatic cystic neoplasm, for which we propose the designation “eosinophilic biliary cystic neoplasm of the liver.” Broader recognition of the unique characteristics of this tumor may facilitate the accumulation of additional cases and further advance understanding of its biological behavior.
Author Contributions
Mariko Tanaka: conceptualization, data analysis and interpretation, and manuscript drafting. Daizo Koinuma: genome sequencing data analysis. Munetoshi Hinata: single‐cell spatial transcriptome data analysis. Kimiko Takeshita: genomic sample acquisition. Yoichi Yasunaga: pathological interpretation. Kei Sakuma: transcriptome sample acquisition. Takeshi Takamoto, Yudai Nakai, and Kiyoshi Hasegawa: clinical data provision. Yudai Nakai: radiological data provision. Tetsuo Ushiku: conceptualization, data interpretation, critical revision, and final approval of the manuscript.
Ethics Statement
The study was approved by the University of Tokyo Ethics Committee (G‐2210) and conducted in accordance with the Declaration of Helsinki.
Consent
Full informed consent was obtained from the patient.
Conflicts of Interest
T.U. serves as an Associate Editor of Pathology International. M.T. is a member of the Editorial Board of Pathology International. The remaining authors declare no conflicts of interest.
Supporting information
Figure S1: Gene set enrichment analysis of RNA sequencing data showed enrichment of the RAS dependency signature in tumor tissue, with a normalized enrichment score of 1.616.
Figure S2: (A) UMAP visualization of the 7 major cell types identified by single‐cell spatial transcriptomic analysis.
Table S1: Lists of immunohistochemical antibodies.
Table S2: Immunohistochemical characteristics of tumor cells.
Table S3: Gene Set Enrichment Analysis results of spatial transcriptomics according to the MsigDB.
Supporting Figure Legends.
Supplementary Data 1: Methods of genome sequencing.
Acknowledgments
This work was supported by JSPS KAKENHI Grant Number JP25K10245.
References
- 1. Nakanuma Y., Basturk O., Esposito I., Klimstra D. S., Komuta M., and Zen Y., Intraductal Papillary Neoplasm of the Bile Ducts. WHO Classification of Tumours of the Digestive System, 5th ed. (International Agency for Research on Cancer (IARC), 2019), 279–282. [Google Scholar]
- 2. Basturk O., Nakanuma Y., Aishima S., et al., Mucinous Cystic Neoplasm of the Liver and Biliary System. WHO Classification of Tumours of the Digestive System, 5th ed. (International Agency for Research on Cancer (IARC), 2019), 250–253. [Google Scholar]
- 3. Torbenson M., Zen Y., and Yeh M. M., “Benign and Borderline Biliary Neoplasms and Tumor‐Like Lesions. AFIP Atlas of Tumor Pathology Series 4, Fascicle 27: Tumor of the Liver,” American Registry of Pathology (2018): 153–199. [Google Scholar]
- 4. Nakanuma Y., Jang K. T., Fukushima N., et al., “A Statement by the Japan‐Korea Expert Pathologists for Future Clinicopathological and Molecular Analyses Toward Consensus Building of Intraductal Papillary Neoplasm of the Bile Duct Through Several Opinions at the Present Stage,” Journal of Hepato‐Biliary‐Pancreatic Sciences 25, no. 3 (2018): 181–187. [DOI] [PubMed] [Google Scholar]
- 5. Singhi A. D., Wood L. D., Parks E., et al., “Recurrent Rearrangements in Prkaca and PRKACB in Intraductal Oncocytic Papillary Neoplasms of the Pancreas and Bile Duct,” Gastroenterology 158 (2020): 573–582.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Vyas M., Hechtman J. F., Zhang Y., et al., “DNAJB1‐PRKACA Fusions Occur in Oncocytic Pancreatic and Biliary Neoplasms and Are Not Specific for Fibrolamellar Hepatocellular Carcinoma,” Modern Pathology 33 (2020): 648–656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Singh A., Greninger P., Rhodes D., et al., “A Gene Expression Signature Associated With “K‐Ras Addiction” Reveals Regulators of Emt and Tumor Cell Survival,” Cancer Cell 15, no. 6 (2009): 489–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jones D. C., Elz A. E., Hadadianpour A., Ryu H., Glass D. R., and Newell E. W., “Cell Simulation as Cell Segmentation,” Nature Methods 22, no. 6 (2025): 1331–1342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hao Y., Stuart T., Kowalski M. H., et al., “Dictionary Learning for Integrative, Multimodal and Scalable Single‐Cell Analysis,” Nature Biotechnology 42, no. 2 (2024): 293–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Tanaka M., Takeshita K., Kunita A., Hasegawa K., and Ushiku T., “PRKACA/PRKACB Fusions in Pancreatobiliary Intraductal Oncocytic Papillary Neoplasms Including Those With Atypical Morphology: An Analysis of 22 Cases Expanding Morphologic Spectrum,” American Journal of Surgical Pathology 48, no. 8 (2024): 1032–1040. [DOI] [PubMed] [Google Scholar]
- 11. Kobayashi T., Omori Y., Ono Y., et al., “Pathways for the Development of Multiple Epithelial Types of Intraductal Papillary Mucinous Neoplasm of the Pancreas,” Journal of Gastroenterology 56, no. 6 (2021): 581–592. [DOI] [PubMed] [Google Scholar]
- 12. Schlitter A. M., Born D., Bettstetter M., et al., “Intraductal Papillary Neoplasms of the Bile Duct: Stepwise Progression to Carcinoma Involves Common Molecular Pathways,” Modern Pathology 27, no. 1 (2014): 73–86. [DOI] [PubMed] [Google Scholar]
- 13. Nakanuma Y., “A Novel Approach to Biliary Tract Pathology Based on Similarities to Pancreatic Counterparts: Is the Biliary Tract an Incomplete Pancreas?,” Pathology International 60, no. 6 (2010): 419–429. [DOI] [PubMed] [Google Scholar]
- 14. Nakanuma Y. and Tsui W. M., Bile Duct Adenoma. WHO Classification of Tumours of the Digestive System, 5th ed. (International Agency for Research on Cancer (IARC)., 2019), 245–247. [Google Scholar]
- 15. Arena V., Arena E., Stigliano E., and Capelli A., “Bile Duct Adenoma With Oncocytic Features,” Histopathology 49, no. 3 (2006): 318–320. [DOI] [PubMed] [Google Scholar]
- 16. Nakanuma Y. and Tsui W. M., Biliary Adenofibroma. WHO Classification of Tumours of the Digestive System, 5th ed. (International Agency for Research on Cancer (IARC)., 2019), 248–249. [Google Scholar]
- 17. Liao X., Agostini‐Vulaj D., Li R. X., and Zhang X., “Characterization of Cholangiocarcinomas With Tubulocystic Morphology Associated With Biliary Adenofibroma or Biliary Adenofibroma‐Like Lesions,” Modern Pathology 38, no. 10 (2025): 100815. [DOI] [PubMed] [Google Scholar]
- 18. Godambe A., Brunt E. M., Fulling K. H., and Reza Kermanshahi T., “Biliary Adenofibroma With Invasive Carcinoma: Case Report and Review of the Literature,” Case Reports in Pathology 2016 (2016): 8068513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Williamson S. R., Hartmann A., Hes O., Martignoni G., Rioux‐Leclercq N. C., and Saleeb R. M., Papillary Renal Cell Carcinoma. WHO Classification of Tumours of the Urinary and Male Genital Tumours, 5th ed. (International Agency for Research on Cancer (IARC), 2022), 47–49. [Google Scholar]
- 20. Crawford J. M., Bioulac‐Sage P., and Hytiroglou P., “Structure, Function, and Responses to Injury,” in MacSween's Pathology of the Liver, eds. Burt A. D., Ferrell L. D., and Hübscher S. G. (Elsevier, 2023), 6th ed., 1–95. [Google Scholar]
- 21. Enomoto K. and Nishikawa Y., “Development and anatomy of the bile duct,” in Pathology of the Bile Ducts, eds. Nakanuma Y. (Springer, 2017), 3–18. [Google Scholar]
- 22. Tabibian J. H., Masyuk A. I., Masyuk T. V., O'Hara S. P., and LaRusso N. F., “Physiology of Cholangiocytes,” Comprehensive Physiology 3, no. 1 (2013): 541–565. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Gene set enrichment analysis of RNA sequencing data showed enrichment of the RAS dependency signature in tumor tissue, with a normalized enrichment score of 1.616.
Figure S2: (A) UMAP visualization of the 7 major cell types identified by single‐cell spatial transcriptomic analysis.
Table S1: Lists of immunohistochemical antibodies.
Table S2: Immunohistochemical characteristics of tumor cells.
Table S3: Gene Set Enrichment Analysis results of spatial transcriptomics according to the MsigDB.
Supporting Figure Legends.
Supplementary Data 1: Methods of genome sequencing.