Introduction

It is often difficult to correctly diagnose a suspicious biliary stricture. Biliary brushes and biopsies usually yield low cellularity and when cytopathological assessment is indecisive, additional sensitive diagnostic tests are lacking. Next generation sequencing (NGS) is an adjunctive diagnostic tool that may improve the diagnostic sensitivity. The aim of this study is to assess the added value of NGS for morphological classification of biliary brushes and biopsies in patients with suspicious biliary strictures.

Aims & Methods

In this retrospective single-center cohort study between 2019-2022, patients with suspicious biliary strictures of which biliary specimens by brush or intraductal biopsy were obtained and on which NGS (Ion Torrent) was performed, were included. Targeted NGS panels covering 40 (v5.1) and 59 genes (v6.1) were used for analysis of biopsies whereas brushes were analysed using the Oncomine™ Colon cfDNA Assay covering 14 genes with >240 hotspots. NGS was classified as malignant by an expert cytopathologist whenever one or more pathogenic mutations were identified. Sensitivity and specificity of NGS were calculated in general and in addition to type of morphology; i.e. benign, atypical, suspicious for malignancy and malignant. Final diagnosis of the suspicious biliary strictures was defined on surgical resection specimens and autopsy, other endoscopic or percutaneous biopsies and/or clinical follow-up (>1 year). In addition, the changes in clinical decision making (CDM) after NGS outcome were evaluated retrospectively by an expert treatment panel. CDM was defined as a change in the course of treatment or the decisive factor to continue current treatment, without the need of additional imaging or pathology material.

Results

A total of 109 samples in 106 patients were included (94 brushes, 15 biopsies). Without taking into account initial morphology/histology, NGS had an overall sensitivity of 65% for brushes and 58% for biopsies, and specificity of 100% for both. Whenever taking initial morphology and histology into account, this resulted in an 85% sensitivity and 93% specificity. Whenever the initial morphology was described as atypical, NGS was able to correctly classify 54% of malignant cases, with a 64% sensitivity and 100% specificity. Overall, NGS was able to identify 42 of the 75 (56%) malignancies correctly. There were no false positive results. Table 1 shows the sensitivities and specificities for both the brush and biopsy groups, stratified by initial morphology. The outcome of NGS resulted in a change of CDM in 8% of the patients.

Conclusion

There is a significant additional yield of NGS in the setting of biliary strictures. Nevertheless, NGS has a limited impact on clinical decision making and for now should only be used in patients in which the outcome is most valuable. This could be to differentiate between different causes of cancer to determine the optimal chemotherapy regimen, or in patients with a clinical suspicion of malignancy but without any mass on cross-sectional imaging. With more sensitive NGS panels for cholangiocarcinoma and more targeted therapy options, in the future the clinical impact will likely increase.

Disclosure

M.J. Bruno received research funding for industry initiated studies from Boston Scientific and Cook Medical. He received research funding for investigator initiated studies from Boston Scientific, Cook Medical, Pentax Medical, Interscope, Mylan and ChiRoStim. He is a consultant to Boston Scientific, Cook Medical, and Pentax Medical. The other authors declare no conflicts of interest.

Introduction

Despite the use of various multidisciplinary diagnostic tools, early differentiation between benign and malignant biliary strictures remains challenging. BILEMUT consists of a mutational analysis of bile cfDNA collected at the time of first ERCP using a next-generation sequencing (NGS) panel of 52 genes (1). It has demonstrated a sensitivity and specificity of 96.4% and 69.2%, respectively, for malignancy in biliary strictures. While genetic analyses were once very expensive, they have become more affordable nowadays. When evaluating new diagnostic techniques, it is crucial to consider not only their accuracy but also their cost, adverse events, and potential for early diagnosis.

Aims & Methods

Prospective observational descriptive pilot study in a cohort of patients diagnosed with non-malignant biliary strictures after the first ERCP recruited in a tertiary hospital between 2017-2020. The aim of this study is to compare the differences in diagnostic costs in euros between performing BILEMUT and repeating tests to achieve a diagnosis according to current clinical practice in biliary strictures. Moreover, adverse events and time required for diagnosis were evaluated.

Results

From the cohort of 68 patients with biliary stricture, 35 received a non-malignant diagnosis after undergoing the first ERCP. Of those patients, 12 were directly referred to surgery, 7 were considered for a wait and see or palliative care approach due to age and/or comorbidity, and 16 required further diagnostic tests to confirm the diagnosis.
In these 16 patients, the median cost of the different endoscopic procedures, radiological techniques, hospital admissions and consultations performed was 8641.52 euros (895.9 – 31704.4). The cost of BILEMUT was 1223 euros per patient. BILEMUT assay had a S=100%, E= 62.5%, NPV=100%, PPV=72,7%. Repeating tests to achieve a definitive diagnosis according to current clinical practice in biliary strictures yield a S = 62.5%, E=100%, PPV=100% and NPV=62.5%. Both strategies had the same diagnostic accuracy in this cohort, 81.3% (57-93.4%). Among the 8 patients who were diagnosed with malignant biliary stricture, the median time to reach the final diagnosis in real clinical practice was 43 days (range 10-930). In contrast, BILEMUT analysis was obtained much faster, with a median turnaround time of 10 days (range 7-14 days). BILEMUT does not pose any additional risks to patients. Furthermore, no complications occurred in the current clinical practice strategy in this cohort despite the extra invasive tests.

Conclusion

This pilot study suggests that the BILEMUT approach in patients with indeterminate biliary strictures after the first ERCP could reduce costs without impairing diagnostic accuracy compared to the current clinical practice approach. In addition, BILEMUT reduced the time and number of diagnostic tests needed to achieve a final diagnosis. There could be a lower theoretical risk of complications associated with the BILEMUT approach due to its ability to reduce the need for invasive tests. However, this study did not demonstrate such a risk reduction.

References

Arechederra M, Rullán M, Amat I, et al. Gut 2022;71:1141–1151.

Introduction

Indeterminate biliary duct strictures (IBDS) pose a diagnostic dilemma, and digital single-operator cholangioscopy (d-SOC) has improved diagnostic accuracy by enabling targeted intra-ductal biopsies (1-3). There is no consensus on the optimal biopsy strategy, but data suggests that at least four single biopsies yields adequate specimen samples for pathological examination (4). Considering the desmoplastic character of cholangiocarcinoma obtaining larger and deeper samples may have a higher diagnostic yield. The aim of this study was to compare the diagnostic yield of d-SOC guided single standard biopsies to those obtained via bite-on-bite-biopsy (BBB) technique in patients with IBDS.

Aims & Methods

This international, multi-center, prospective cohort study included patients with a diagnosis of IBDS who underwent d-SOC from November 2020 to August 2022 at fourteen European tertiary referral centers (Dutch Trial Register: NL9649). IBDS was defined as a bile duct stricture of indeterminate nature after previous laboratory work-up, imaging or endoscopic retrograde cholangiography with or without brush-cytology of fluoroscopy guided intra-ductal biopsies. During d-SOC, in every patient sampling of the stricture(s) was performed by firstly obtaining at least 4 single biopsies, and secondly obtaining at least 1 BBB. Definitive diagnosis was established on pathology outcomes (biopsies or surgical resection specimens) and clinical follow-up of at least one year. Primary outcome was the accuracy of both biopsy techniques.

Results

Eighty-nine patients were included (62% male, median age: 66 years), 52 of whom had a hilar and 37 a distal stricture. Single and BBB biopsies were technically successful and with sufficient tissue for diagnosis in 82 (92.1%, median number of biopsies = 4) and 78 (87.6%, median number of BBB biopsies = 2), respectively. These biopsies confirmed malignancy in 31/82 and 29/78 of cases, respectively. Comparing the two biopsy techniques, these techniques yielded different results in 4/78 patients (5.1%).
In 82 (92.1%) patients follow-up was complete and malignancy was confirmed in 51 (62.2%) patients, resulting in an overall sensitivity, specificity and accuracy of 60.8%, 100% and 75.6% of both techniques combined. For both sampling techniques, sensitivity and accuracy decreased significantly if a stent was placed at a prior ERC (n = 37, 41.6%) or whenever prior intra-ductal tissue acquisition had been performed (n = 41, 46.1%). The number of BBB did not affect sensitivity or accuracy. No adverse events related to d-SOC guided biopsies were noted.

Conclusion

In this prospective study, BBB did not outperform at least four random single biopsies of IBDS. Prior manipulation of the IBDS, by stent placement or prior tissue acquisition, is associated with a decreased yield.

References

1. Angsuwatcharakon P, Kulpatcharapong S, Moon JH, et al. Consensus guidelines on the role of cholangioscopy to diagnose indeterminate biliary stricture. HPB (Oxford). 2022;24(1):17-29.
2. Navaneethan U, Hasan MK, Lourdusamy V, et al. Single-operator cholangioscopy and targeted biopsies in the diagnosis of indeterminate biliary strictures: a systematic review. Gastrointestinal Endoscopy. 2015;82(4):608-614.e602.
3. Almadi MA, Itoi T, Moon JH, et al. Using single-operator cholangioscopy for endoscopic evaluation of indeterminate biliary strictures: results from a large multinational registry. Endoscopy. 2020;52(7):574-582.
4. Bang JY, Navaneethan U, Hasan M, et al. Optimizing Outcomes of Single-Operator Cholangioscopy–Guided Biopsies Based on a Randomized Trial. Clinical Gastroenterology and Hepatology. 2020;18(2):441-448.e441.

Disclosure

P.J.F. de Jonge is a speaker and received a consultancy fee for Boston Scientific. J.J. Vila is a speaker for Pentax, Cook Medical, and Olympus and is a consultant for Boston Scientific. M.W. James received consultancy fees from Boston Scientific and Cook Medical. K.W. Oppong received research funding for investigator initiated studies from Medtronic, and received speaker fees from Medtronic and Boston Scientific. D. Joshi is a speaker and received consultancy fees from Boston Scientific, Cook Medical, and Mirum Pharmaceuticals. M. Ellrichmann received research funding for industry initiated studies from Boston Scientific. He received research funding for investigator initiated trials from Boston Scientific, He is a consultant to Boston Scientific, Microtech, Medwork and Olympus. L. Kylänpää received a speaker fee from Boston Scientific. F. van der Heide received a speaker fee from AbbVie. P. Hindryckx is a speaker and received consultancy fees from Medtronic, Boston Scientific, Viatris, Fujifilm, and Medwork. V. Cennamo is a consultant and speaker for and received travel grants from Olympus Italia, Olympus Europe, and Euromedical. G.J.M. Webster is a consultant and speaker for Boston Scientific, Cook Medical, Pentax Medical, and Olympus Europe. M.J. Bruno received research funding for industry initiated studies from Boston Scientific and Cook Medical. He received research funding for investigator initiated studies from Boston Scientific, Cook Medical, Pentax Medical, Interscope, Mylan, and ChiRoStim. He is a consultant to Boston Scientific, Cook Medical, and Pentax Medical. The remaining authors have no conflicts of interests to disclose.

Introduction

Diagnosis of biliary obstructions is a challenge especially when histopathology is inconclusive. Indeterminate biliary strictures often require additional verification and a personalized approach.

Aims & Methods

This study investigates the prevalence, characteristics, and natural history of indeterminate biliary strictures. This retrospective study covers patients treated with ERCP because of jaundice due to biliary stricture(s) of unknown etiology. Exclusion criteria were prior liver transplantation, visible tumor, lithiasis, acute pancreatitis, PSC, or any known cause of dilated biliary ducts with subsequent biliary stricture diagnosed with imaging studies (US, CT, MRI, or EUS) and serology and clinical data. A diagnosis of potentially benign biliary stricture was based on cytology/histopathology evaluation of samples taken on primary ERCP. The study participants were monitored every 2–3 months with further endoscopic, imaging, and serological evaluations as needed.

Results

A total of 510 patients were treated with primary ERCP for biliary obstruction of indeterminate etiology. The 186 (36.5%) had biliary strictures with negative cytology/histopathology (category II per Papanicolaou guidelines), of these 55 (29.6%) had strictures in the liver hilum, 22 (11.8%) in the common bile duct (CBD), and 108 (58.1%) in the intrapancreatic region. Malignancy was confirmed in subsequent tests in 40 patients (21.5%). A benign cause was diagnosed in 77 cases (41.4%), including pancreatitis, PSC, ischemia, IgG4-SC, and Mirizzi syndrome, while 69 strictures (37.1%) remained indeterminate. Strictures in the liver hilum had a higher rate of eventual malignancy (30%) compared to CBD (23%) and intrapancreatic (16%) localization. Lithiasis was more common in intrapancreatic strictures (12%), while PSC was more frequent in hilar strictures (14%). During 6-month follow-up of the 69 indeterminate cases: 17 patients received a diagnosis (8 lithiasis, 5 malignancies, 2 chronic pancreatitis, 1 benign neoplasm, 1 IgG4-SC), 4 patients died from unrelated causes, 48 cases (25.8% of the potentially benign stricture group) remained indeterminate.

Conclusion

A quarter of benign biliary strictures remain indeterminate in 6 months follow-up, while 20% of patients with indeterminate receive diagnosis of malignancy.

Introduction

Differentiating malignant from benign biliary obstruction is critical to avoid delayed cancer diagnosis or unnecessary invasive procedures. In many developing countries, diagnostic pathways are often limited or delayed. While serum total bilirubin is widely accessible, its predictive potential for malignancy remains underutilized.

Aims & Methods

This study aimed to evaluate the diagnostic performance of total bilirubin in distinguishing benign from malignant biliary obstruction and to propose a practical cut-off for early triage, especially in settings lacking immediate access to advanced imaging or endoscopy.
A retrospective study was conducted at a tertiary care center between May 2016 and October 2024, using a prospectively maintained ERCP database. Two groups of 450 patients each—with confirmed benign or malignant obstruction—were randomly selected. Diagnoses were based on histology, imaging, or consistent clinical follow-up. Total serum bilirubin, measured prior to any biliary drainage, was expressed in mg/L. Comparative analyses used Student’s t-test, and ROC curve analysis identified the optimal threshold via the Youden Index.

Results

In the benign group (mean age 65.0 years, 29.8% male), etiologies included 397 cases of common bile duct stones, 21 hydatid cysts with biliary fistula, and 32 benign strictures. The mean total bilirubin was 59.3 mg/L.
In the malignant group (mean age 64.1 years, 49.8% male), etiologies included 173 pancreatic tumors, 126 cholangiocarcinomas (hilar or extrahepatic), 93 ampullomas, and 58 cases of metastatic or contiguous invasion. The mean total bilirubin was 201.4 mg/L.
The difference between groups was highly significant (p<0.000001). ROC analysis showed an AUC of 0.88, indicating excellent diagnostic performance. The Youden Index identified a threshold of 104 mg/L, yielding 81.8% sensitivity and 84.7% specificity. Above this threshold, the probability of malignancy increased nearly fivefold (OR 4.6; 95% CI 3.5–6.1; p<0.0001).

Conclusion

Total serum bilirubin shows excellent accuracy in distinguishing malignant from benign biliary obstruction. A threshold of 104 mg/L could serve as a simple, low-cost triage tool for prioritizing patients with indeterminate strictures, especially in resource-limited settings. Integrating bilirubin thresholds into early diagnostic algorithms may expedite oncologic care and improve patient outcomes.

Introduction

By definition, indeterminate biliary strictures are those that cannot be diagnosed with routine imaging modalities, including ultrasound, CT scan, and MRI. In such cases, single-operator cholangioscopy (SOC) has emerged as a useful tool for diagnostic and therapeutic interventions, which enables visualization of the bile duct and allows for targeted biopsies of suspicious lesions. The use of SOC has been shown to improve diagnostic accuracy, decrease the need for additional procedures, and guide therapeutic interventions. Nevertheless, pooled sensitivity for diagnosis of SOC-guided biopsies in the diagnosis of malignant biliary strictures is limited with approximately 60%. So far, there is no evidence on which biopsy technique is best and how many biopsy samples need to be taken to improve diagnostic accuracy.

Aims & Methods

Aim: To evaluate the best sampling technique comparing two versions of the SpyBite forceps to optimize diagnostic yield for the differentiation of indeterminate biliary stricture.
Patients and methods: Patients with an indeterminate biliary stricture with an independent indication to perform an ERCP with SOC and bile duct sampling were recruited to participate in this prospective, randomized, multicenter study. Patients were equally randomized into the following study arms: (i) single biopsy (SB); (ii) multiple biopsies with suction technique (MB); (iii) bite-on-bite biopsies (BB). At least 6 biopsies had to be obtained in each group. Furthermore, the best sampling technique was transferred into phase 2 of the study comparing conventional forceps (Spybite) vs. larger forceps (Spybite max.). Final diagnosis was defined as prove of malignancy, histology of surgical resectate or 6 months follow-up without signs of malignancy. Adequacy of the samples was rated according to a 6 point rating scale (cytological samples rate 0-2; histological samples rates 3-5). Sensitivity/specificity of samples were defined as primary endpoint.

Results

A total of 60 patients (20pt. per group, mean age 67y, 48% female) were included in the first study phase. In the SB, MB and BB group a mean of 6.4, 6.7 and 6.5 samples was taken, respectively. Of note, in the MB group a significant difference of the number of samples being evaluated (∆=3.5; -47.8%) was observed indicating a loss of samples during suction. Resulting sensitivity/specificity of the obtained samples was: SB 76/89%; MB 61/85%; BB 79/92%, with a significant lower level in the MB group compared to SB and BB (p=0.01). In the 2^nd study phase 20 patients were included per group. Usage of the larger forceps significantly increased sensitivity to 85% (p=0.01) with a specificity of 95% applying the BB technique. Conventional forceps obtained cytological samples in 48.3% and histological samples in 51.7%, whereas larger forceps resulted in 8.3% cytological samples (p=0.0001) and histological samples in 91.7% (p=0.001).

Conclusion

Based on the results of this study, the bite-on-bite biopsy technique using larger forceps (Spybite max.) appears to be the best sampling technique for optimizing diagnostic yield in the differentiation of indeterminate biliary stricture.

Disclosure

Mark Ellrichmann, Arne Kandulski, Markus Dollhopf, Arno Dormann, Alexander Dechene, Torsten Beyna, Christian Gerges received lectures fees from Boston Scientific