The core problem is anatomical and biological. Cholangiocarcinoma is classified into intrahepatic, perihilar, and distal subtypes, each with distinct growth behavior. Intrahepatic tumors may appear as mass-forming lesions, as periductal infiltrating growth that creeps along the bile duct wall, or as papillary tumors growing inside the duct lumen. Perihilar and distal tumors, which make up the majority of cases in referral centers, typically present as infiltrative sclerosing strictures with longitudinal subepithelial spread rather than as discrete masses. This morphological diversity means that a technique excellent at sampling one subtype may be nearly useless for another. Moreover, the tumor is characterized by abundant desmoplastic stroma, a dense fibrotic reaction composed of cancer-associated fibroblasts, inflammatory cells, and a collagen-rich extracellular matrix that often constitutes a substantial proportion of the tumor volume. The result is tissue with relatively few malignant cells, which explains why conventional sampling techniques so often return inadequate or nonrepresentative specimens.

The numbers tell a sobering story. A landmark meta-analysis by Navaneethan and colleagues found that the pooled sensitivity of brush cytology for detecting malignant biliary strictures is only about 45 percent, while intraductal forceps biopsy reaches roughly 48 percent. Combining both methods raises sensitivity to only about 59 percent, even though specificity remains excellent at approximately 99 percent. In practical terms, this means that a negative cytology or histology result cannot reliably exclude malignancy, forcing clinicians to repeat procedures or pursue multimodal evaluation before committing to surgery or systemic therapy. Fluorescence in situ hybridization, which detects chromosomal abnormalities associated with biliary malignancy, can improve matters considerably; adding FISH to brush cytology and biopsy has achieved diagnostic sensitivities approaching 82 percent in selected populations. But the field has been waiting for techniques that yield higher-quality tissue and greater diagnostic confidence from the outset.

Enter endoscopic ultrasound, or EUS, which has evolved over three decades from a purely imaging modality into a comprehensive diagnostic and interventional platform. Because the echoendoscope sits in close proximity to the biliary system, it can detect subtle pathological changes invisible on computed tomography or magnetic resonance imaging, particularly in small distal lesions and early-stage disease. Research shows that EUS outperforms CT and MRCP in distinguishing malignant from benign biliary strictures, with reported sensitivities ranging from 25 to 91 percent and specificities from 89 to 100 percent. Beyond imaging, EUS enables real-time tissue acquisition from primary tumors, regional lymph nodes, and metastatic lesions in a single procedure, and studies have reported up to 85 percent accuracy in assessing vascular invasion, a critical factor in determining whether a tumor can be surgically resected. The introduction of fine-needle biopsy needles, which provide core tissue with preserved architecture rather than mere cytology, has been transformative: these specimens are suitable not only for histopathological examination but also for immunohistochemistry and next-generation sequencing.

That last point matters enormously in the era of precision oncology. Comprehensive genomic profiling has revealed that cholangiocarcinoma is a molecularly heterogeneous disease harboring several actionable alterations, including FGFR2 fusions, IDH1 mutations, HER2 amplification, BRAF V600E mutations, NTRK fusions, and microsatellite instability. Targeted therapies directed against these biomarkers have markedly expanded treatment options for selected patients, and international guidelines now recommend molecular profiling in unresectable or metastatic disease whenever adequate tissue is available. EUS-guided fine-needle biopsy has become an important platform for this work, with several studies demonstrating that it yields a higher proportion of samples suitable for complete genomic profiling while reducing the need for repeat tissue acquisition. In other words, the same needle that establishes the diagnosis can also determine which drug the patient should receive.

But EUS has a blind spot: infiltrative tumors that do not form a solid, defined mass are difficult to identify and sample under ultrasound guidance, particularly in the perihilar region. This is where digital single-operator cholangioscopy enters the picture. Systems such as the SpyGlass DS allow the endoscopist to steer a tiny camera directly into the bile duct, visualizing the mucosal surface in real time. Characteristic features suggestive of malignancy include irregular mucosal nodularity, papillary projections, ulceration, and dilated tortuous vessels, with abnormal neovascularization being one of the most reliable visual indicators. A meta-analysis by de Oliveira and colleagues, encompassing 283 patients undergoing digital single-operator cholangioscopy, demonstrated a pooled sensitivity of 94 percent and specificity of 95 percent for visual impression alone. In patients with prior nondiagnostic ERCP-based sampling, visual assessment achieved sensitivities of 90 to 100 percent and overall diagnostic accuracy approaching 97 percent. Cholangioscopy-directed biopsy, meanwhile, achieves tissue adequacy in about 93 percent of cases with sensitivity around 75 percent and specificity of 100 percent according to a large multinational registry.

The Bulgarian team, led by Petko Karagyozov, Nadica Male, Kalina Panova, and Stela Nikolova at Acibadem City Clinic University Hospital Tokuda in Sofia, put these complementary techniques to the test in a retrospective single-center analysis of 117 consecutive patients evaluated for suspected biliary malignancy between January 2020 and September 2025. Following surgical histopathology or multidisciplinary consensus, 96 patients, or 82 percent, were confirmed to have malignant disease, with perihilar lesions predominating at 78 percent. EUS-guided fine-needle biopsy was performed in 40 patients with confirmed malignancy, achieving a sensitivity of 65 percent, a figure the authors attribute to the predominance of infiltrative hilar lesions lacking a clear sonographic target. Cholangioscopy-directed biopsy was performed in 72 malignant cases, achieving a sensitivity of 72.2 percent, while characteristic cholangioscopic visual features were identified in 61 of those patients, corresponding to a sensitivity of 84.7 percent.

The most striking result emerged in the subgroup of 26 patients who underwent both procedures. When evaluated individually, each modality demonstrated only moderate sensitivity, but combining the two approaches increased overall sensitivity to 96.2 percent, with only a single malignant lesion remaining undiagnosed after the combined endoscopic evaluation. This finding aligns with previous work: Lee and colleagues reported a combined sensitivity of 100 percent for EUS-FNA plus cholangioscopy-guided biopsy, while Moura and colleagues achieved 86 percent with EUS-FNA plus ERCP-based sampling. Both procedures in the Bulgarian cohort demonstrated an acceptable safety profile, with mild pancreatitis occurring in 4.5 percent of patients after EUS and cholangitis in 13.5 percent after cholangioscopy, all classified as mild and resolved with conservative treatment. No severe adverse events or procedure-related mortality were observed.

The authors are careful to frame these findings as exploratory and hypothesis-generating rather than definitive. The study was retrospective and single-center, the choice of diagnostic modality was individualized according to clinical judgment rather than a standardized protocol, and the combined-modality subgroup comprised only 26 patients. Formal paired statistical comparisons between the two techniques were not performed, and detailed procedural characteristics such as needle type and gauge were not consistently available. Nevertheless, the cohort reflects real-world clinical practice at a tertiary referral center, precisely the setting where diagnostic uncertainty is greatest and multimodal evaluation is most likely to provide meaningful benefit. The authors emphasize that the choice of endoscopic approach should be tailored to lesion morphology and anatomical location: patients with discrete mass lesions or suspected nodal disease benefit most from EUS, whereas infiltrative strictures without an identifiable mass are better evaluated with cholangioscopy.

Looking ahead, the diagnostic landscape is evolving rapidly beyond the endoscope itself. Artificial intelligence and machine learning algorithms are being developed to assist in interpreting EUS images and identifying malignant cholangioscopic features, potentially reducing interobserver variability and supporting real-time decision-making. Probe-based confocal laser endomicroscopy enables real-time microscopic assessment of the biliary epithelium during ERCP, while radiomics and radiogenomic approaches are extracting quantitative imaging biomarkers from CT, MRI, and EUS examinations. Liquid biopsy, analyzing circulating tumor DNA, circulating tumor cells, microRNAs, and extracellular vesicles, offers the promise of minimally invasive diagnosis and longitudinal monitoring of treatment response, particularly valuable in patients with limited tissue availability. The ultimate vision, according to the authors, is an integrated diagnostic algorithm that combines high-resolution imaging, endoscopic visualization, targeted tissue acquisition, molecular profiling, and computational decision support, tailored to each patient’s tumor morphology, anatomical location, and molecular characteristics. For a cancer that has resisted diagnosis for decades, the message is clear: no single lens is enough, but two lenses looking from different directions may finally bring the tumor into focus.

Subject of Research: Multimodal endoscopic diagnosis in biliary oncology: integrating EUS-guided tissue acquisition and cholangioscopy for personalized cancer care

Article Title: Multimodal endoscopic diagnosis in biliary oncology: integrating EUS-guided tissue acquisition and cholangioscopy for personalized cancer care

Article References: Karagyozov, P., Male, N., Panova, K., & Nikolova, S. (2026). Multimodal endoscopic diagnosis in biliary oncology: integrating EUS-guided tissue acquisition and cholangioscopy for personalized cancer care. Frontiers in Gastroenterology, 5, Article 1924535. https://doi.org/10.3389/fgstr.2026.1924535

Image Credits: AI Generated

DOI: 10.3389/fgstr.2026.1924535

Keywords: cholangiocarcinoma, endoscopic ultrasound, cholangioscopy, precision oncology, biliary cancer, diagnostic endoscopy, molecular profiling, Multimodal, endoscopic, diagnosis, biliary, oncology