Turning discarded bile into a source of genomic information
Distinguishing malignant from benign biliary strictures remains a major clinical challenge. Endoscopic retrograde cholangiopancreatography (ERCP)-guided cytology and biopsy may be inconclusive, particularly in early-stage disease or when tissue is scarce. Serum carbohydrate antigen 19-9 (CA19-9) also has limited specificity because cholestasis and inflammation can substantially increase its concentration.
A study published in eGastroenterology suggests that bile collected during routine ERCP could provide an additional source of tumour-derived genomic information. Rojo and colleagues evaluated ultra-low-pass whole-genome sequencing (ULP-WGS) of bile cell-free DNA as a potentially scalable and cost-efficient approach for malignancy detection, tumour-origin classification and prognostic stratification.
Prospective analysis of 95 biliary strictures
The study included 95 patients undergoing their first ERCP for a biliary stricture radiologically suspicious for malignancy. Initial pathological assessment classified 26 strictures as malignant and 69 as indeterminate. After up to 12 months of follow-up, 45 initially indeterminate cases were confirmed as malignant and 24 as benign. The final malignant cohort comprised 40 cholangiocarcinomas (CCAs) and 31 pancreatic ductal adenocarcinomas (PDACs).
Using only 2 ng of bile cell-free DNA, the researchers performed ULP-WGS at approximately 0.2× genomic coverage. The open-source ichorCNA pipeline was used to identify large-scale copy-number alterations (CNAs) and estimate the tumour fraction (TF), with TF ≥10% indicating detectable circulating tumour DNA.
High specificity and earlier detection
Among patients with an initial pathological diagnosis of malignancy, bile ULP-WGS detected tumour DNA in 73% of cases, including 82% of CCAs and 56% of PDACs.
In the more clinically challenging group of initially indeterminate strictures that were subsequently confirmed as malignant, sensitivity was 51%. Importantly, 23 of 24 ultimately benign cases tested negative, corresponding to a specificity of 96%. By comparison, CA19-9 at the prespecified threshold of 44 U/mL achieved only 53% specificity in evaluable benign cases.
In the 23 ULP-WGS-positive patients whose malignancy was not established at the first ERCP, tumour DNA was detected a median of 38 days before conventional diagnosis. In seven patients, the genomic signal preceded diagnosis by more than 100 days, with the longest interval reaching 499 days.
Targeted bile sequencing using the investigators' 52-gene Bilemut assay remained more sensitive, detecting mutations in 94.7% of evaluable malignant strictures. ULP-WGS, however, required less DNA and provided genome-wide CNA information that targeted mutation analysis could not capture.
Identifying the probable tumour origin
The CNA profiles reproduced genomic alterations previously described in pancreatobiliary tumour tissue, including recurrent losses of 9p, 17p and 18q and gains of 7p, 7q, 8q and 19q.
A supervised 50-region CNA classifier distinguished CCA from PDAC with an accuracy of 86.4% in the reference cohort and 85.0% in a separate validation cohort. Loss of 1p was enriched in PDAC, whereas 9p loss occurred more frequently in CCA. These findings suggest that the integrated CNA pattern, rather than a single alteration, may encode tissue-of-origin information.
Genomic integration may improve risk stratification
Neither TF nor targeted tumour mutational burden (TMB) alone significantly stratified overall survival. However, patients who were both TF-positive and had mutations in at least three genes had a median overall survival of 5.1 months, compared with 13.7 months among the remaining patients.
After adjustment for initial management, CA19-9 and carcinoembryonic antigen, this combined genomic profile remained independently associated with poorer survival (adjusted hazard ratio 2.51, 95% CI 1.22–5.14).
Potential clinical role
Bile ULP-WGS is not sufficiently sensitive to exclude malignancy when the result is negative and should not replace histopathology. Its most promising role may be as a complementary test after an inconclusive first ERCP. A positive result could justify accelerated repeat sampling, cholangioscopy, endoscopic ultrasound-guided biopsy or closer multidisciplinary surveillance.
Reducing sequencing depth from conventional 30–50× WGS to 0.2× was estimated to lower sequencing costs by approximately 130–215-fold. Nevertheless, the single-centre cohort was modest, the benign group was small and the CCA–PDAC classifier requires external validation. Prospective multicentre studies must determine whether this strategy can reduce repeated procedures, shorten diagnostic delays and ultimately improve patient outcomes.
See the articles:
Rojo C, Rullan M, Sogbe M, et al. Ultralow-pass whole-genome sequencing of bile cfDNA in patients with biliary strictures. eGastroenterology 2026;4:e100425. doi:10.1136/ egastro-2026-100425
About eGastroenterology
eGastroenterology, a BMJ journal partnered with Gut and launched by leading scientists in gastroenterology and hepatology, has been indexed in the Web of Science Core Collection (ESCI), PubMed, DOAJ, Scopus, CAS, ROAD, and many other major international databases within just two years of its launch.
eGastroenterology has recently received its first Journal Impact Factor of 10.5 and now ranks 11/153 in the Web of Science Gastroenterology & Hepatology category.
