Exploring the biliary microbiome in hepatopancreatobiliary disorders: a comprehensive systematic review of microbial signatures and diagnostic potential Authors: Hossein Halimi, Zeinab Hesami, Nastaran Asri, Babak Khorsand, Mohammad Rostami-Nejad, Hamidreza Houri Published: 2025 ABSTRACT AbstractBackgroundHepatopancreatobiliary (HPB) diseases, encompassing hepatobiliary and pancreatic disorders, pose substantial global health challenges due to their high morbidity and mortality rates. Recent research highlights the crucial role of the biliary microbiome in the development of these diseases.MethodsThis study provides a comprehensive systematic review of the biliary microbiome’s characteristics across various HPB disorders, including cholangiocarcinoma (CCA), pancreatic cancer (PC), primary sclerosing cholangitis (PSC), and gallstone disease (GSD). Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we screened articles from multiple databases, focusing on original research utilizing 16 S rRNA gene sequencing or metagenomics.ResultsOur review included 24 studies that met stringent inclusion criteria. The results indicate distinct alterations in bacterial diversity and composition associated with different HPB conditions, highlighting potential pathogenic mechanisms and candidate taxa as potential microbial indicators. In lithiasis conditions, elevated levels of Pyramidobacter and Citrobacter were associated with recurrent and giant common bile duct (CBD) stones. Proteobacteria were prevalent in PSC and CCA, potentially contributing to these diseases by promoting chronic inflammation. Sphingomonas was associated with both CCA and PSC, with potential implications for lymph node metastasis in PC.ConclusionsThese findings suggest the potential of the biliary microbiome as a diagnostic tool, offering insights into the pathophysiology and possible therapeutic targets for HPB diseases. However, given the heterogeneity in methodologies and the limited number of studies including healthy controls, these observations remain preliminary; further prospective validation is required before clinical translation.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12876-025-04551-1.Keywords: Biliary microbiome, Cholangiocarcinoma, Biliary lithiasis, Pancreatic cancer, And primary sclerosing cholangitis Introduction The complex network of hepatopancreatobiliary (HPB) diseases and associated malignancies encompasses a spectrum of gastrointestinal diseases, including hepatobiliary and pancreatic disorders. Hepatobiliary disease stands as a significant contributor to global morbidity and mortality, accounting for 1.472 million deaths worldwide in 2019, a significant rise from 1.013 million deaths recorded in 1990 [1, 2]. Within the spectrum of hepatobiliary diseases, gallstone disease (GSD) presents a significant public health concern on a global scale, with prevalence rates spanning from 2% to 20% among various populations. Biliary stones are the leading cause of hospital admissions and surgical procedures, placing a considerable burden on healthcare resources [3]. Cholangiocarcinoma (CCA), a rare form of hepatobiliary cancer with a poor outlook, exhibits varying incidence rates ranging from 85 to 0.3 per 100,000 individuals in northeast Thailand and Israel, respectively [4]. Primary sclerosing cholangitis (PSC) is marked by inflammation and fibrosis of the bile ducts, representing a significant predisposing factor for CCA [5]. The persistent inflammation within the bile ducts creates an environment conducive to tumorigenesis, a phenomenon often referred to as multifocal carcinogenesis [6]. As indicated by the Global Burden of Disease (GBD) study, pancreatic diseases play a substantial role in global morbidity and mortality. Pancreatic cancer (PC), in particular, stands out as one of the most lethal malignancies, characterized by high mortality rates [7]. Frequently diagnosed in advanced stages, PC presents a dismal prognosis. PC cases are escalating, and projections indicate it will rank as the second leading cause of cancer mortality by 2030 [8, 9]. Accumulating evidence substantiates the fundamental influence of the human microbiome on HPB pathophysiology [10–13]. Notably, the intricate interplay between the gut microbiota and bile acids has catapulted interest, with significant changes in gut microbiota composition observed in individuals suffering from various liver and gallbladder ailments. Despite the exponentially growing interest in the relationship between the gut microbiome and pancreaticobiliary disorders, the precise impact of the biliary microbiome on disease etiology remains elusive, primarily due to challenges associated with sample acquisition [14]. Nevertheless, recent years have witnessed concerted efforts aimed at elucidating the enigmatic role of the bile microbiome within the HPB system, challenging the traditional notion of bile as a sterile environment. In this vein, Molinero et al. delve into an exploration of the bile microbiome in healthy controls, revealing the presence of predominant bacterial phyla such as Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria [15]. In our study, we conducted a comprehensive and systematic review of the existing literature to investigate the changes in the biliary microbiome associated with HPB diseases. This research represents the first systematic exploration of the implications of the bile microbiota across a spectrum of HPB disorders, including CCA, PC, PSC, and biliary stone diseases. We hypothesized that the specific bacterial communities in bile could serve as potential risk factors for the development and progression of various HPB conditions, and how could apply these alterations in the biliary microbiome could be applied to differentiate benign conditions from HPB malignancies as a potential prognostic screening tool. Methods Study design This systematic review was conducted following the rigorous methodology outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to explore the relationship between bile microbiota and HPB diseases [16]. To ensure transparency and reproducibility, the study protocol was prospectively registered on PROSPERO (CRD42024526433), and the completed PRISMA checklist is provided in Supplementary File 1. Search strategy In our search for relevant articles related to our study focus, we systematically explored electronic databases including MEDLINE (via PubMed), Web of Science, Scopus, Embase, and Cochrane, employing a carefully curated set of keywords. These keywords encompassed a broad spectrum of HPB disorders, such as “Primary sclerosing cholangitis”, “Cholangitis”, “Primary biliary cholangitis”, “Primary biliary Cirrhosis” “Cholestasis”, “Pancreatic cancer”, “Pancreatic ductal adenocarcinoma”, “Liver disease”, “Hepatoma”, “Hepatocellular carcinoma”, “Cholangiocarcinoma”, “Bile duct cancer”, “Hepatoblastoma”, “Malignant liver”, “Liver angiosarcoma”, “Hepatic adenomas”, “Lithiasis”, “Choledocholithiasis”, “Cholelithiasis”, “Gallstone”, and “Common bile duct stone”, “Gallbladder stone”. Simultaneously, we specifically targeted terms related to the biliary microbiome, such as “Biliary microbiome”, “Biliary microbiota”, “Biliary bacteria”, “Bile microbes”, “Bile microbiome”, and “Bactobilia”. This search strategy spanned from the inception of the databases up to April 2025. Two investigators, H.H. and Z.H., precisely assessed both abstracts and full-text articles, adhering closely to the predefined inclusion and exclusion criteria established for this study, under the supervision of HR.H. Furthermore, we conducted cross-referencing of retrieved articles to identify any potentially ignored research contributions. Eligibility criteria Our study adhered to a set of precisely defined eligibility criteria to ensure the reliability and validity of our conclusions. Articles were carefully reviewed to ensure they met a comprehensive set of criteria. These criteria included the presence of original research on the bile microbiome, with a particular focus on a variety of HPB diseases such as CCA, PC, and PSC, as well as conditions associated with stone formation. Additionally, stringent requirements were established on the source of samples, necessitating their recruitment of bile fluid. Moreover, the methodology employed for microbiome identification was restricted to 16 S rRNA gene sequencing or metagenomics techniques. Our analytical approach demanded comprehensive assessments, including evaluations of alpha and beta diversity, relative abundance variations, and the identification of potential microbial indicators in the microbiome. Moreover, our study outlined exclusion criteria aimed at filtering out research that could compromise the integrity of our analysis. Articles falling under the categories of reviews, editorials, or book chapters were excluded from data synthesis to ensure the focus remained on primary original research investigations. Similarly, studies conducted exclusively on non-human samples were deemed ineligible for inclusion, thus maintaining the human-centric focus of our analysis. Additionally, methodologies relying on culture or quantitative PCR techniques were excluded from consideration, given their potential limitations in capturing the complex dynamics of the bile microbiome. Furthermore, studies that lacked distinct comparative analyses between groups or focused solely on microbial abundance without suitable comparison were excluded. This selection criterion properly ensured the exploration of the factors influencing the bile microbiome in HPB diseases. Quality assessment of the included studies We assessed the quality and methodological reliability of the chosen observational studies using the widely accepted Newcastle–Ottawa Scale (NOS). This scale assesses three pivotal factors: population selection, study comparability, and exposure. Scores on the scale span from 0 to 9, with ratings of 0–4 indicating poor quality, 5–7 suggesting moderate quality, and scores surpassing seven signifying good quality. Only studies achieving a NOS score of ≥ 7 were deemed eligible for inclusion in our analysis. Assessment of study quality was conducted independently by our research team members, H.H. and Z.H. In cases of discrepancy, consultation with our supervisor, HR.H., was sought to reconcile disparities and establish a final consensus. Result Study selection Our initial search across PubMed, Web of Science, Scopus, Embase, and Cochrane databases, targeting articles investigating the relationship between the biliary microbiome composition and PC, CCA, PSC, and lithiasis, yielded a total of 1566 records. Following the elimination of duplicate articles, non-English publications, and irrelevant titles/abstracts, our refined dataset comprised 320 articles. Subsequently, a detailed and rigorous evaluation of the full-text content of each article against our predetermined inclusion and exclusion criteria led to the identification of 24 articles eligible for inclusion in our final assessment using the Newcastle-Ottawa Scale. The PRISMA flow diagram detailing our selection process is presented in Fig. 1. PRISMA diagram of studies identified in this systematic review Summary of included studies The general characteristics and details of the included studies are summarized in Table 1. All the studies obtained bile samples during the endoscopic retrograde cholangiopancreatography (ERCP) procedure. In terms of laboratory methods, all studies utilized 16 S rRNA gene sequencing, with one study also incorporating metagenomics alongside 16 S rRNA gene sequencing. The majority of studies targeted the V3-V4 region of the 16 S rRNA gene, though some focused on the V4, V3, V1-V2, and V1-V3 regions to detect the bacterial microbiome. The most commonly used pipelines for data analysis were QIIME and DADA2, while the Silva and Greengenes databases were frequently employed for taxonomic classification. Operational taxonomic units (OTUs) and amplicon sequence variants (ASVs) were used to classify bacteria based on sequence similarity. Table 1General characteristics and details of the included studiesAuthor (year)Study TypeCountryTarget populationFemale (%)Median or mean age inyearsPrimary OutcomeMicrobial analysismethodComputational methodsTarget RegionInclusion criteriaExclusion criteriaQualityRatingPoud et el. (2023) [33]Case-controlUSACCA (n = 6), PC (n = 25)11 (34.4)66 (37–82)Malignancy-related microbiomefingerprints in bile samples of benign and malignant pancreaticobiliarydiseases.16 S rRNA gene Metagenomic SequencingQIIME, DADA2, SilvaV3-V4--ModerateMiyabe et al. (2022) [37]Case-controlUSACCA (n = 24), Lithiasis (n = 5)6 (25)/3 (60)62 (51–73)/69 (59–69)Both bile and stool microbiota profiles in PSC and CCA16 S rRNA gene SequencingGreengenes, RDP-Persistently elevated alkaline phosphatase levels for at least six months, with subsequent confirmation of a bile-duct stricture by MRCP or ERCPIndividuals with secondary sclerosing cholangitis and benign biliary stricturesModerateLi et al. (2022) [27]Case-controlChinapCCA (n = 14), dCCA (n = 9), PC (n = 8), cholelithiasis (n = 22)5 (35.7)/4 (44.4)/4 (50.0)/16 (72.7)68.1 ± 6.0/61.6 ± 4.3/58.6 ± 10.6/60.3 ± 14.7Bacterial biomarker of CCA and PC16 S rRNA geneSequencingQIIME, SilvaV4Volunteers aged 43–80 years oldIndividuals with other systemic diseases, recent use of proton pump inhibitors, antibiotics, prebiotics, or other medical procedures; those with a history of neoplastic diseases; and those with oral ulcers in the past few monthsHighDangtakot et al. (2021) [36]Case-controlThailandCCA (n = 30), choledocholithiasis (n = 31)9(30)/12(38)61.26 ± 10.38/64.73 ± 14.71Bacterial relative abundance and biomarker of patients with CCA16 S rRNA geneSequencingQIIME, UCHIME, SilvaV3-V4--ModerateSaab et al. (2021) [39]Case-controlIraneCCA (n = 28), cholelithiasis (n = 47)9 (32)/24 (51)64/57Relative abundance of bile microbiota in eCCA16 S rRNA geneMetagenomic SequencingFLASH2, ENA, SilvaV3-V4Individuals aged 21 years or older suspected of biliary obstruction due to cholangiocarcinoma or lithiasis, indicated by elevated liver enzyme and total bilirubin levels, and a stricture observed on EUS or MRCPIndividuals with autoimmune hepatitis, metabolic hepatitis, viral hepatitis, alcoholic hepatitis, NASH, chemotherapy, and medication-induced hepatitisHighChen et al. (2019) [26]Case-controlChinadCCA (n = 8), choledocholithiasis (n = 44)3(37)/26(59)-Relative abundance and bacterial biomarker of dCCA and recurrent CBD stone16 S rRNA genepyrosequencing--Diagnosed with CCA via CT, MRI, or abdominal ultrasoundIndividuals with previous cancer diagnosis, PSC, CBD, Caroli disease, and CCAModerateAvilés-Jiménez et al. (2016) [38]Case-controlMexicoeCCA (n = 10), control (CBD stone(N = 8),normal biliary duct (n = 1),benign stenosis of choledocus (n = 1))5(50)/8(80)Age-matched (± 5 years)Relative abundance of biliary microbiota in eCCA and the detection of Helicobacter pylori cagA and vacA genes16 S rRNA genesequencingPHYCA-STATS, NCBI, GreengenesV4--HighLiwinski et al. (2020) [42]Case-controlGermanyPSC (n = 43), choledocholithiasis and papillary adenoma (n = 22)16(43)/11 (50)39 (20–55)/55 (22–89)Microbiomes of the upper alimentary tract and ductal bile in PSC patients, in addition to measuring bile acid16 S rRNA geneSequencingDADA2, SilvaV1-V2PSC diagnosis confirmed by typical biliary lesions on cholangiography and liver biopsyIndividuals with secondary causes of sclerosing cholangitis, acute bacterial cholangitis during the initial ERC, patient age, and a previous ERC within the past 12 monthsModeratePereira et al. (2017) [43]Case-controlFinlandPSC (n = 69), control (n = 46)25 (68)/17 (50)/27 (59)38 (14)/41 (13)/43 (15)Biliary microbiota in PSC patients and its duration16 S rRNA geneSequencingEuropean Nucleotide Archive (ENA)V1-V3--ModerateMerali et al. (2023) [34]Case-controlUKPDAC (n = 12), control n = 17 (CBD stone(N = 10), gallstone (N = 6), PSC(N = 1))5 (41%)/5 (30%)71 (47–93)/64 (33–84)Role of bile microbiota in PDAC patients16 S rRNA geneSequencingDADA2, GTDBV3-V4Individuals aged 18 years or older with a WHO performance status of 0, 1, or 2, willing to provide written informed consent, presenting with a biliary stricture, and confirmed benign or malignant pancreaticobiliary disease during the initial ERCP or PTBDIndividuals with no clinical or image data about pancreaticobiliary disease; pregnant women, patients undergoing ERCP post-bariatric surgery, hepatico-jejunostomy, or Bilroth II surgery; any infectious disease such as hepatitis or HIVModerateLiu et al. (2024) [28]Case-controlChinaRecurrent (n = 27), non-recurrent choledocholithiasis (n = 29)11(40)/13(44)69.04 ± 13.24/58.41 ± 17.09Relative abundance and biomarkers of biliary microbiota in the recurrent stone formation group, in addition to its metabolic functions16 S rRNA geneSequencingQIIMEV3-V4Non-recurrent patients without a history of ERCP and surgical procedures; recurrent patients whose stones were removed more than 6 months agoPatients aged < 18 years, severe cardiopulmonary complications, recent antibiotic treatment within six months before enrollment, secondary bile stones, incomplete stone removal, abdominal malignancy, congenital biliary malformation, cirrhosis, pancreatitis, and gastrointestinal tract reconstructionModeratePark et al. (2024) [23]Case-controlKoreaCholelithiasis (n = 11),choledocholithiasis (n = 17),8(72)/9(52)59 (30–81)/67 (27–88)Relative abundance and biomarker evaluation of lithiasis patients16 S rRNA geneSequencingVSEARCH, EzBioCloudV3-V4Absence of acute cholangitis or acute cholecystitis at the time of diagnosis, no history of ERCP or biliary surgery, no antibiotic therapy within the past three months, and not currently using probiotics-HighCai et al. (2023) [17]Case-controlChinaCholelithiasis (n = 31),choledocholithiasis (n = 15), healthy (n = 8)6(40)/14(45)/3 (37)39.71 ± 12.37/56.03 ± 16.97/32.37 ± 10.34Bacterial biomarkers in lithiasis patients and the impact of bacteria on outcomes following endoscopic surgery16 S rRNA gene novaseqDADA/QIIMEV3-V4Age between 18–80 years and symptomatic gallstone diseasePatients unable to provide informed consent, individuals with biliary tract or liver disease comorbidities outside the study criteria, those with asymptomatic gallstones, participants with a history of ERCP or other biliary surgeries, and those who received antibiotic therapy within the past three monthsModerateDai et al. (2023) [21]Case-controlChinaGiant (n = 8) and small Choledocholithiasis (n = 18)2(25), 5(27)83.50 ± 6.28/66.67 ± 14.48Bacterial biomarker evaluation on patients with large and small biliary stones16 S rRNA genenovaseqQIIME, RDPV3-V4Diagnosed with CBD through abdominal ultrasound, CT, MRCP, or EUS; no antibiotic use within the preceding three monthsIndividuals with a history of surgery involving the upper gastrointestinal tract or common bile duct, symptoms of acute obstructive suppurative cholangitis, and hepatobiliary-pancreatic neoplastic diseasesHighLee et al. (2023) [25]Case-controlKoreaCholedocholithiasis (n = 17), control(dilated bile duct (n = 4), a benign tumor (n = 1), benign biliary stricture (n = 1), CBD bile sludge (n = 1), and autoimmune pancreatitis (n = 1))9 (52)/3 (37)72 (43–89)/71 (42–81)Biliary microbiome in patients with CBD stones alongside their metabolic profiles16 S rRNA geneSequencingQIIME2, GreengenesV3-V4Adults aged ≥ 18 years diagnosed by CT or MRI, and ERCP are includedPatients with severe liver disease or hemolytic anemia, a history of challenging endoscopic procedures, and a medical history of thrombocytopenia or coagulopathyModerateLiu et al. (2023) [18]Case-controlChinaCholedocholithiasis (n = 33), control (n = 17)15(45)/8(47)65.00 (13.73)/59.94 (13.58)Biliary bacterial abundance, microbiota-derived amino acid metabolites, and the secondary bile acid-related pathway in patients with CBD stones16 S rRNA geneSequencingQIIME, DADA2, SilvaV3-V4-Patients using antibiotics, prebiotics, probiotics, or symbiotics in the prior two months, and those with secondary choledocholithiasis, hepatolithiasis, pancreatitis, or hepatobiliary tumors.HighLi et al. (2023) [30]cohortChinaRecurrent (n = 4), non-recurrent choledocholithiasis (n = 39)1 (25.00)/15 (38.46)44 (38–46)/47 (38–64)Biliary microbiome abundance and risk factors in the recurrent stone formation group, in addition to metabolic pathways related to this disease16 S rRNA geneSequencingGreengenesV3-V4-Individuals with a history of malignant diseases, autoimmune diseases, diabetes, any exposure to antibiotics within the past month, or structural abnormalities of the biliary tractHighTan et al. (2022) [29]Case-controlChinaRecurrent (n = 5), non-recurrent choledocholithiasis (n = 45)--Biliary bacterial microbiome abundance and biomarker comparison between patients with recurrent stone formation and non-recurrent cases16 S rRNA geneSequencingQIIMEV3-V4None of the patients had received antibiotic treatment within three months before the ERCP procedure-ModerateFeng et al. (2022) [22]Case-controlChinaCholelithiasis (n = 14)choledocholithiasis (n = 18), control (N = 13)(bile duct stenosis (n = 5), PC (n = 3), pancreatic mass (n = 2), Diverticulum (n = 1), congenital bile duct abnormality (n = 1), bile duct malignancy (n = 1))7(50)/11(61)/7(53)62.28/63.11/61.38Biliary bacterial biomarker of lithiasis group, as well as surveying a related metabolic pathway16 S rRNA geneSequencingQIIME, DADA2, SilvaV1-V3No patients had received antibiotic treatment within three months before the ERCP procedure-HighLyu et al. (2021) [24]Case-controlChinaCholedocholithiasis (n = 15), control (pancreatic without stone) (n = 4))5 (33)/0 (0)68.13± 13.53/39.25 ± 12.09Biliary and duodenal bacterial composition in CBD stones and related metabolic pathways16 S rRNA geneSequencingQIIME, SilvaV3-V4Patients diagnosed with choledocholithiasis but without hepatobiliary diseases, and without severe conditions affecting the heart, brain, kidneys, etc., and in good health, underwent ERCP after providing signed informed consentAntibiotic use within three months before sample collection, coexisting cholelithiasis, and malignant tumors of the bile ductModerateChoe et al. (2021) [19]CohortKoreaRecurrent (n = 11), non-recurrent choledocholithiasis (n = 9)4 (44)/5 (45)62 (32–78)/63 (29–83)Biliary bacterial profile and bile composition of a group with recurrent stone formation16 S rRNA geneSequencingFLASH, NCBI, GreengenesV3-V4Patients who had no biliary tract infection, did not use antibiotics or proton pump inhibitors for at least three months before ERCP, and had stones larger than 3 mm detected after six months are considered recurrent patients-HighKim et al. (2021) [31]Case-controlKoreaCholedocholithiasis (n = 13), control (n = 15) (benign biliary stricture (n = 4), gallbladder cancer (n = 4), PC (n = 3), distalbiliary tract cancer (n = 3), or hepatocellular carcinoma (n = 1)9 (69)/8 (53)72 (22–84)/69 (51–83)Biliary bacterial biomarkers in patients with CBD stones16 S rRNA geneSequencingSilvaV3-V4Diagnosed by abdominal ultrasound, CT, MRI, and ERCP, with a naïve ampullaInability to provide written informed consent, use of pancreatic ERCP, previous procedures involving the ampulla, cholesterol stone, recurrent CBD stone after ERCP, and challenging endoscopic approachModerateYe et al. (2020) [20]Case-controlChinaRecurrent choledocholithiasis (n = 27), control(N = 17) (oddi dysfunction (n = 3), secondary cholangitis (n = 5), surgical injury (n = 1), PDAC (n = 3),4 CCA (n = 4), ampullary carcinoma (n = 1))10(37)/9(52)62 (34–90)/53 (14–78)Biliary bacterial abundance in relation with the recurrent stone formation group16 S rRNA geneSequencingQIIME, GreengenesV3-V4Diagnosed by CT or MRCP, with a history of stone removal six months agoPatient age < 18 years, severe cardiopulmonary complications, CBD stone could not be removed or the stone originated from the gallbladder, and a previous biliary procedure within the last six monthsModerateMolinero et al. (2019) [15]Case-controlSpainCholelithiasis (n = 14), healthy control (n = 13)11 (78)/9 (69)Age range 27 to 73/aged range 37 to 79Biliary bacterial abundance and bile metabolism related to lithiasis in both healthy and affected individuals16 S rRNA gene SequencingQIIME, DADA2, Silva, NCBIV3Adults under 80 years old who stayed at the ICU 24 and 48 h prior to sample collection, respectively, and did not receive antibiotics-ModeratePC pancreatic cancer, CCA cholangiocarcinoma, PSC primary sclerosing cholangitis, pCCA perihilar cholangiocarcinoma, dCCA distal cholangiocarcinoma, eCCA extrahepatic CCA ENA European Nucleotide Archive, CBD common bile duct, PDAC pancreatic ductal adenocarcinoma, MRCP magnetic resonance cholangiopancreatography, ERCP endoscopic retrograde cholangiopancreatography, EUS endoscopic ultrasound, PTBD percutaneous transhepatic biliary drainage, WHO World Health Organization, NASH non-alcoholic steatohepatitis; CT computed tomography, MRI magnetic resonance imaging, ICU intensive care unit General characteristics and details of the included studies 66 (37– 82) Malignancy-related microbiome fingerprints in bile samples of benign and malignant pancreaticobiliary diseases. 5 (35.7)/ 4 (44.4)/ 4 (50.0)/ 16 (72.7) 68.1 ± 6.0/ 61.6 ± 4.3/ 58.6 ± 10.6/ 60.3 ± 14.7 16 S rRNA gene Sequencing 9(30)/ 12(38) 16 S rRNA gene Sequencing 16 S rRNA gene Metagenomic Sequencing 16 S rRNA gene pyrosequencing eCCA (n = 10), control (CBD stone (N = 8), normal biliary duct (n = 1), benign stenosis of choledocus (n = 1)) 16 S rRNA gene sequencing 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing Cholelithiasis (n = 11), choledocholithiasis (n = 17), 16 S rRNA gene Sequencing Cholelithiasis (n = 31), choledocholithiasis (n = 15), healthy (n = 8) 16 S rRNA gene novaseq Choledocholithiasis (n = 17), control (dilated bile duct (n = 4), a benign tumor (n = 1), benign biliary stricture (n = 1), CBD bile sludge (n = 1), and autoimmune pancreatitis (n = 1)) 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing Cholelithiasis (n = 14) choledocholithiasis (n = 18), control (N = 13) (bile duct stenosis (n = 5), PC (n = 3), pancreatic mass (n = 2), Diverticulum (n = 1), congenital bile duct abnormality (n = 1), bile duct malignancy (n = 1)) 16 S rRNA gene Sequencing 68.13 ± 13.53/39.25 ± 12.09 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing Choledocholithiasis (n = 13), control (n = 15) (benign biliary stricture (n = 4), gallbladder cancer (n = 4), PC (n = 3), distal biliary tract cancer (n = 3), or hepatocellular carcinoma (n = 1) 16 S rRNA gene Sequencing 16 S rRNA gene Sequencing PC pancreatic cancer, CCA cholangiocarcinoma, PSC primary sclerosing cholangitis, pCCA perihilar cholangiocarcinoma, dCCA distal cholangiocarcinoma, eCCA extrahepatic CCA ENA European Nucleotide Archive, CBD common bile duct, PDAC pancreatic ductal adenocarcinoma, MRCP magnetic resonance cholangiopancreatography, ERCP endoscopic retrograde cholangiopancreatography, EUS endoscopic ultrasound, PTBD percutaneous transhepatic biliary drainage, WHO World Health Organization, NASH non-alcoholic steatohepatitis; CT computed tomography, MRI magnetic resonance imaging, ICU intensive care unit Selecting appropriate healthy control groups presented a significant challenge due to the inherent difficulties in obtaining bile samples from healthy individuals and the limitations of current methodologies for analyzing bile with low bacterial load. Consequently, cancer subjects were frequently compared with individuals suffering from biliary lithiasis or other benign hepatobiliary conditions. Two studies included samples from healthy controls without hepatobiliary diseases obtained during liver transplants from liver donors who had suffered a brain accident or stroke. Controls were thus variably defined across studies as healthy donors (e.g., liver transplant recipients without HPB disorders), patients with benign biliary conditions (e.g., lithiasis, strictures, or cholangitis), or other non-malignant HPB disorders (e.g., pancreatitis or congenital abnormalities). This variability may affect interpretations by introducing potential biases, such as confounding microbiome alterations due to underlying benign conditions, antibiotic exposure in transplant settings, or procedural differences, which could obscure disease-specific microbial signatures. Geographically, the studies were conducted in China (n = 11), Korea (n = 4), the USA (n = 2), Thailand (n = 1), Mexico (n = 1), Iran (n = 1), Finland (n = 1), Spain (n = 1), the UK (n = 1), and Germany (n = 1). Biliary microbiome in patients with lithiasis The composition of the biliary microbiome in cases of lithiasis may be linked to chronic inflammation, potentially contributing to the development of HPB cancer. To explore this potential link, we categorized 15 stone-related articles into three groups: (i) a comparison between lithiasis and controls, which included 7 articles involving 166 patients with common bile duct stones or gallstones, and 78 control participants, including those without hepatobiliary disorders (particularly liver donors) and those with the benign biliary disease without stones (e.g., pancreatitis, bile duct stricture, benign bile duct abnormalities, etc.); (ii) comparison between gallstones and common bile duct stones, which included two articles involving 25 patients with gallstones and 35 with common bile duct stones; and (iii) comparison between recurrent and non-recurrent stone formation, which included 6 articles involving 90 patients with recurrent stone formation and 183 in the non-recurrent group. This categorization allowed us to systematically evaluate the relationship between biliary lithiasis and various hepatobiliary conditions, providing insights into how chronic inflammation from stone formation may contribute to cancer development. Biliary microbial diversity in lithiasis In the context of biliary lithiasis, alterations in bacterial diversity are evident when compared to a control group (Table 2). Furthermore, disparities in bacterial diversity are noted between patients with common bile duct (CBD) stones and those with GSD, as well as between recurrent stone formation and non-stone formation groups. Taken together, these findings unveil distinct alterations in bacterial diversity correlated with different conditions in biliary lithiasis, thereby underscoring its potential implications for HPB health. Figure 2 illustrates the alpha and beta diversity of the biliary microbiome in patients with various lithiasis. These findings shed light on the intricate interplay between the biliary microbiome and lithogenic processes, offering valuable insights into the pathophysiology of biliary diseases and potential avenues for targeted interventions. Table 2Characteristics of the biliary Microbiome alterations associated with biliary lithiasis: microbial abundance, diversity, and biomarkers across reviewed studiesBiliary LithiasisArticleCase/ControlEnriched AbundanceDepleted AbundancePotential BomarkerAlpha DiversityBeta DiversityFeng et al. [22]Bile stone/Non-stone--↑= f-alchaligenaceae↓= f-bradyrhizobiaceae, f-lachnospiraceae, o- Burkholderiales, o-rhizobiales--Feng et al. [22]Cholelithiasis/Choledocholithiasis--↑= f-Propionibacteriaceae, f-Sphingomonadaceae, f-Lactobacillaceae, f-Verrucomicrobiaceae (p = 0.001), g-Campylobacter(p = 0.007), s-Propionibacterium acnes (p = 0.013), s-Peptoclostridium difficileObserved OTU: Sig higher (p = 0.015)-Liu et al. [18]Choledocholithiasis/Benign biliary disease-p-Actinobacteriota (p = 0.000), p-Patescibacteria (p = 0.004), p-Spirochaetota (p = 0.005), p-Candidatus_Saccharibacteria (p = 0.01), p-Cyanobacteria (p = 0.04)-Observed OTU: Sig lower (p = 0.00012)Shannon: Sig lower (p = 0.00018)Simpson: Sig lower (p = 0.0046)Chao1: Sig lower (p = 0.00005)Pielou: Sig lower (p = 0.021)Unweighted Unifrac: Sig (p = 0.001)Bray-Curtis: Sig (p = 0.01)Weighted Unifrac: Sig (p = 0.04)Kim et al. [31]Choledocholithiasis/Non-stone--↑= g-Enterococcus (p = 0.031)↓= p-Bacteroidetes_unclassified o-Clostridiales_unclassifiedChao1: NSShannon: NS-Ye et al. [20]Recurrent Choledocholithiasis/Non-stonep-proteobacteria (p = 0.037)f-Enterococcaceaeg-Escherichiap-Bacteroidetes, p-actinobacteria (p = 0.001)f-Oxalobacteraceae, f-Moraxellaceae, f-Caulobacteraceae, f-Burkholderiaceae-Observed OTU: NSChao1: NSShannon: Sig lowerSimpson: Sig lowerWeighted Unifrac: SigTan et al. [29]Recurrent Choledocholithiasis/Non-recurrent Choledocholithiasisp-synergistetes, p-proteobacteria, g-klebsiellap-bacteroidetes, p-actinobacteria↑= g-Kurthia, g-Paraclostridium, g-Enterobacter, g-Klebsiella, g-Raoultella, g-Pyramidobacter↓= p-Spirochaetes, p-Bacteroideteso-Corynebacteriales, o-Bacteroidales, o-Chitinophagales, o-Caulobacterales, o-Betaproteobacteriales, o-Pseudomonadales, o-Xanthomonadales, o-Spirochaetales,f-Nocardiaceae, f-Micrococcaceae, f-Prevotellaceae, f-Chitinophagaceae, f-Flavobacteriaceae, f-Caulobacteraceae, f-Burkholderiaceae, f-Moraxellaceae, f-Xanthomonadaceae, f-Spirochaetaceae,g-Rhodococcus, g-Porphyromonas, g-Alloprevotella, g-Prevotella, g-Brevundimonas, g-Caulobacter, g-Novosphingobium, g-Curvibacte, g-Acinetobacter, g-Stenotrophomonas--Li al. [30]Recurrent Choledocholithiasis/Non-recurrent Choledocholithiasiso-Pasteurellales, o-Neisseriales, o-Campylobacterales, g-Veillonella, g-Oribacterium, g-Neisseria, g-Leptotrichia, g-Campylobactero-Pseudomonadales, o-Burkholderiales, o-Bacillales, o-Aeromonadales, o-Actinomycetales,g-Enterococcus, g-Clostridium, and an unclassified genus of Aeromonadaceae↑= o-Fusobacteriales (p < 0.05) o-Neisseriales (p < 0.05)↓= o-LactobacillalesFisher alpha: NScanonical correspondence analysis; Sig (p = 0.03)Choe et al. [19]Recurrent Choledocholithiasis/Non-recurrent Choledocholithiasisp-Actinobacteria(p = 0.048) p-Firmicutes(p = 0.038)p-bacteroidetes(p = 0.042), g-Neisseria (p = 0.008), g-Capnocytophaga(p = 0.025), g-Gemella(p = 0.041)-Observed OTU: Sig lower (p = 0.000)Chao1: Sig lower (p = 0.005)Shannon: Sig lower (p = 0.027)Inverse Simpson: Sig lower (p = 0.036)Unweighted Unifrac: SigLyu et al. [24]Choledocholithiasis/Pancreatitisg-Brevundimonas, g- Prevotella_1(p = 0.037)p-Actinobacteria (p = 0.007), p-Saccharibacteria (p = 0.005), p-Deinococcus-Thermus (p = 0.023),g-Actinomyces (p = 0.000), g-Proteus (p = 0.000), g-Clostridiumsensu_stricto (p = 0.000), g-Klebsiella (p = 0.001), g-Actinobacillus (p = 0.000), g-Lachnospiraceae_UCG-008 (p = 0.002), g-Butyrivibrio (p = 0.002), g-Roseburia (p = 0.003), g-Porphyromonas (p = 0.01), g-Streptococcus (p = 0.012), g-Helicobacter (p = 0.018), g-Enterobacter (p = 0.039)-Chao1: NSACE: NSSimpson: NSShannon: NS-Dai et al. [21]Giant choledocholithiasis/Small Choledocholithiasis--↑= g-Enterococcus, g-Citrobacter, g-Lactobacillus, g-Pyramidobacter, g-Bifidobacterium, and g-Shewanella↓= g-Granulicatella, g-FusobacteriumChao1: NSObserved OTU: NSShannon: NSSimpson: NSWeighted Unifrac: SigChen et al. [26]Recurrent Choledocholithiasis/Non-recurrent Choledocholithiasis--↑= f-Enterococcaceae, f-Brevibacteriaceae, g-Pyramidobacter, Enterococcus, g-Citrobacter, g-Brevibacterium↓= p-Bacteroidetes, f-Aeromonadaceae, f-Prevotellaceae, o-Aeromonadales, o-Rhizobiales, o-Bacteroidales, c-Bacteroidia, c-Alphaproteobacteriag-Nesterenkonia, g-Ralstonia, g-Alloprevotella, g-Aeromonas, g-PrevotellaObserved OTU: NSChao1: NSSimpson: NSShannon: NS-Park et al. [23]Cholelithiasis/Choledocholithiasisg-Enterococcus, g-Lactobacillus, g-Streptococcusp-Proteobacteria (p = 000), f-Enterobacteriaceae (p = 0.000), g-Prevotella (p = 0.024), g-Bacteroides (p = 0.048), g-Ralstonia↑= c-Bacilli, o-LactobacillalesObserved OTU: Sig higher (p = 0.041)Chao1: NSACE1: NSShannon: Sig higher (p = 0.002)Simpson: Sig lower (p = 0.001)Phylogenetic: Sig higher-Molinero et al. [15]Cholelithiasis/Healthy Controlp-Bacteroidetes,f-Bacteroidaceae (p = 0.002), porphyromonadaceae (p = 0.008), prevotellaceae (p = 0.001), veillonellaceae (p = 0.002), g-bacteroides (p = 0.001), g-dialister (p = 0.002), g-escherichia-shigella (p = 0.001)f-Beijerinckiaceae (p = 0.0006), f-Bradyrhizobiaceae(p = 0.001), f-caulobacteraceae (p = 0.002), f-chitinophagaceae(p = 0.004), f-comamonadaceae (p = 0.009), f-Haliangiaceae(p = 0.006), f-Methylobacteriaceae(p = 0.001), f-propionibacteriaceae(p = 0.006), f-sphingomonadaceae(p = 0.001), g-acidibacter(p = 0.005), g-bradyhizobium(p = 0.004), g-brevundimonas(p = 0.003), g-methylobacterium(p = 0.003), g-sphingomonas(p = 0.001)-Shannon: Sig lower (p = 0.038)Weighted UniFrac: SigCai et al. [17]Choledocholithiasis/Healthy Control--↑= g-prevotella, g-klebsiella, s-streptococcus anginosus, s-morganella morganii, s-arthrobacter↓= g-RalstoniaChao1: Sig lower-Cai et al. [17]Cholelithiasis/Healthy Control--↑= p-bdellovibrionota, g-rothia, g-peptostreptococcus, g-serratia, s-bifidobacterium bifidum--Liu et al. [28]Recurrent Choledocholithiasis/Non-recurrent Choledocholithiasisg-fusobacterium, g-enterococcus, g-escherichia, and g-klebsiellag-streptococcus, g-granulicatella, g-prevotella, g-gemmiger, g-ruminococcus↑= g-Enterococcus↓= g-streptococcus, g-granuicatella, g-prevotella, g-actinomyces, g-abiotrophiaShannon: NSChao1: NSBray-Curtis: Sig (p < 0.001)Lee et al. [25]Choledocholithiasis/Benign biliary disease--↑= f- Enterococcaceae, g- Enterococcus.↓= c- Clostridia, c- Betaproteobacteria, o- Bacillales, o- Clostridiales, o- Burkholderiales, f-Streptococcaceae, f- Carnobacteriaceae, f-lactobacillaceae, f-staphylococcaceae, f-veillonellaceae, f-Comamonadaceae, g-Staphylococcus, g-granulicatella, g-lactobacillus, g-streptococcusObserved ASV: Sig lower (p < 0.01)Shannon: Sig lower (p < 0.01)Bray-Curtis: Sig (p = 0.01)operational taxonomic unit, ASV amplicon sequence variant, CBD common bile duct, GSD gallstone disease, NS not-significant, Sig significantThe symbols “↓” and “↑” indicate a decrease and an increase in diversity, respectively Characteristics of the biliary Microbiome alterations associated with biliary lithiasis: microbial abundance, diversity, and biomarkers across reviewed studies ↑= f-alchaligenaceae ↓= f-bradyrhizobiaceae, f-lachnospiraceae, o- Burkholderiales, o-rhizobiales Observed OTU: Sig lower (p = 0.00012) Shannon: Sig lower (p = 0.00018) Simpson: Sig lower (p = 0.0046) Chao1: Sig lower (p = 0.00005) Pielou: Sig lower (p = 0.021) Unweighted Unifrac: Sig (p = 0.001) Bray-Curtis: Sig (p = 0.01) Weighted Unifrac: Sig (p = 0.04) ↑= g-Enterococcus (p = 0.031) ↓= p-Bacteroidetes_unclassified o-Clostridiales_unclassified Chao1: NS Shannon: NS p-proteobacteria (p = 0.037) f-Enterococcaceae g-Escherichia p-Bacteroidetes, p-actinobacteria (p = 0.001) f-Oxalobacteraceae, f-Moraxellaceae, f-Caulobacteraceae, f-Burkholderiaceae Observed OTU: NS Chao1: NS Shannon: Sig lower Simpson: Sig lower ↑= g-Kurthia, g-Paraclostridium, g-Enterobacter, g-Klebsiella, g-Raoultella, g-Pyramidobacter ↓= p-Spirochaetes, p-Bacteroidetes o-Corynebacteriales, o-Bacteroidales, o-Chitinophagales, o-Caulobacterales, o-Betaproteobacteriales, o-Pseudomonadales, o-Xanthomonadales, o-Spirochaetales, f-Nocardiaceae, f-Micrococcaceae, f-Prevotellaceae, f-Chitinophagaceae, f-Flavobacteriaceae, f-Caulobacteraceae, f-Burkholderiaceae, f-Moraxellaceae, f-Xanthomonadaceae, f-Spirochaetaceae, g-Rhodococcus, g-Porphyromonas, g-Alloprevotella, g-Prevotella, g-Brevundimonas, g-Caulobacter, g-Novosphingobium, g-Curvibacte, g-Acinetobacter, g-Stenotrophomonas o-Pseudomonadales, o-Burkholderiales, o-Bacillales, o-Aeromonadales, o-Actinomycetales, g-Enterococcus, g-Clostridium, and an unclassified genus of Aeromonadaceae ↑= o-Fusobacteriales (p < 0.05) o-Neisseriales (p < 0.05) ↓= o-Lactobacillales Observed OTU: Sig lower (p = 0.000) Chao1: Sig lower (p = 0.005) Shannon: Sig lower (p = 0.027) Inverse Simpson: Sig lower (p = 0.036) p-Actinobacteria (p = 0.007), p-Saccharibacteria (p = 0.005), p-Deinococcus-Thermus (p = 0.023), g-Actinomyces (p = 0.000), g-Proteus (p = 0.000), g-Clostridiumsensu_stricto (p = 0.000), g-Klebsiella (p = 0.001), g-Actinobacillus (p = 0.000), g-Lachnospiraceae_UCG-008 (p = 0.002), g-Butyrivibrio (p = 0.002), g-Roseburia (p = 0.003), g-Porphyromonas (p = 0.01), g-Streptococcus (p = 0.012), g-Helicobacter (p = 0.018), g-Enterobacter (p = 0.039) Chao1: NS ACE: NS Simpson: NS Shannon: NS ↑= g-Enterococcus, g-Citrobacter, g-Lactobacillus, g-Pyramidobacter, g-Bifidobacterium, and g-Shewanella ↓= g-Granulicatella, g-Fusobacterium Chao1: NS Observed OTU: NS Shannon: NS Simpson: NS ↑= f-Enterococcaceae, f-Brevibacteriaceae, g-Pyramidobacter, Enterococcus, g-Citrobacter, g-Brevibacterium ↓= p-Bacteroidetes, f-Aeromonadaceae, f-Prevotellaceae, o-Aeromonadales, o-Rhizobiales, o-Bacteroidales, c-Bacteroidia, c-Alphaproteobacteria g-Nesterenkonia, g-Ralstonia, g-Alloprevotella, g-Aeromonas, g-Prevotella Observed OTU: NS Chao1: NS Simpson: NS Shannon: NS Observed OTU: Sig higher (p = 0.041) Chao1: NS ACE1: NS Shannon: Sig higher (p = 0.002) Simpson: Sig lower (p = 0.001) Phylogenetic: Sig higher p-Bacteroidetes, f-Bacteroidaceae (p = 0.002), porphyromonadaceae (p = 0.008), prevotellaceae (p = 0.001), veillonellaceae (p = 0.002), g-bacteroides (p = 0.001), g-dialister (p = 0.002), g-escherichia-shigella (p = 0.001) ↑= g-prevotella, g-klebsiella, s-streptococcus anginosus, s-morganella morganii, s-arthrobacter ↓= g-Ralstonia ↑= g-Enterococcus ↓= g-streptococcus, g-granuicatella, g-prevotella, g-actinomyces, g-abiotrophia Shannon: NS Chao1: NS ↑= f- Enterococcaceae, g- Enterococcus. ↓= c- Clostridia, c- Betaproteobacteria, o- Bacillales, o- Clostridiales, o- Burkholderiales, f-Streptococcaceae, f- Carnobacteriaceae, f-lactobacillaceae, f-staphylococcaceae, f-veillonellaceae, f-Comamonadaceae, g-Staphylococcus, g-granulicatella, g-lactobacillus, g-streptococcus Observed ASV: Sig lower (p < 0.01) Shannon: Sig lower (p < 0.01) operational taxonomic unit, ASV amplicon sequence variant, CBD common bile duct, GSD gallstone disease, NS not-significant, Sig significant The symbols “↓” and “↑” indicate a decrease and an increase in diversity, respectively Alterations in biliary microbial alpha (a) and beta (b) diversities across various hepatopancreatobiliary disorders compared to control groups. Abbreviations: PC, pancreatic cancer; dCCA, distal cholangiocarcinoma; eCCA, extrahepatic cholangiocarcinoma; PSC, primary sclerosing cholangitis; CBD, common bile duct stone; RCBD, recurrent common bile duct stone; GSD, gallstone disease; HS, healthy subject Alpha diversity in lithiasis Across various comparisons, including lithiasis versus non-lithiasis, GSD versus CBD stones, and recurrent stone formation versus non-recurrent cases, notable differences in alpha diversity indices were observed (Fig. 2A). Based on the Chao1 index, research by Cai et al. showed a notably lower diversity in individuals with biliary lithiasis compared to healthy individuals [17]. Liu et al. demonstrated significantly lower Chao1 in patients with CBD stones compared to a control group affected by benign biliary stricture [18]. Choe et al. reported reduced Chao1 in the recurrent CBD stones group compared to the non-recurrent group [19]. In contrast, Ye et al. [20] and Dai et al. [21] found no significant differences in Chao1 between recurrent and non-recurrent groups or giant and small CBD stones, respectively. Feng et al. [22]., Park et al. [23], and Lyu et al. [24] also reported no significant differences in Chao1 index across their comparisons. Based on observed OTU/ASV, a significant decrease in observed OTU was noted in GSD patients compared to those with CBD stones by Feng et al. [22]. Park et al. reported lower observed OTUs in GSD patients compared to CBD stones [23]. Liu et al. showed significantly lower observed OTUs in patients CBD stones versus controls [18]. Choe et al. observed lower observed OTU in recurrent CBD stones versus non-recurrent [19]. Lee et al. found significantly lower observed ASV in patients with CBD stones compared to controls [25]. However, others reported no significant differences in observed OTU across their groups [20, 21, 24, 26]. According to the Shannon index, Molinaro et al. observed higher diversity in healthy controls compared to GSD [15]. Park et al. reported a lower Shannon index in GSD patients compared to patients with CBD stones [23]. Ye et al. observed a lower Shannon index in recurrent patients compared to non-recurrent controls [20]. Choe et al. reported reduced Shannon in recurrent CBD stones versus non-recurrent [19]. Liu et al. demonstrated significantly lower Shannon in patients with choledocholithiasis versus controls [27]. Moreover, Lee et al. found significantly lower Shannon in patients with choledocholithiasis versus controls [25]. In contrast, others have described no significant differences in Shannon across their comparisons [21, 24, 28]. Based on the Simpson index, Park et al. reported a higher diversity in the GSD group compared to the choledocholithiasis group [23]. Ye et al. observed a lower diversity in patients with recurrent choledocholithiasis compared to non-recurrent patients [20]. Choe et al. reported reduced inverse Simpson in recurrent CBD stones versus non-recurrent [19]. Furthermore, Liu et al. demonstrated significantly lower Simpson in choledocholithiasis patients versus controls [28]. Beta diversity in lithiasis Variations in beta diversity metrics further revealed the microbial changes in biliary lithiasis (Fig. 2B). Based on the Bray-Curtis distance, Lee et al. found significant differences in Bray-Curtis values between patients with choledocholithiasis and controls [25]. Liu’s study revealed significant differences in Bray-Curtis distance between recurrent and non-recurrent CBD stones cases [28]. Liu et al. reported significant differences in Bray-Curtis between patients with choledocholithiasis and controls [18]. On the basis of Weighted/Unweighted UniFrac, Molinaro et al. observed significant differences in Weighted UniFrac between healthy controls and GSD [15]. Ye et al. noted significant differences in Weighted UniFrac between recurrent and non-recurrent [20]. Choe et al. reported remarkable differences in Unweighted UniFrac between recurrent and non-recurrent CBD stones [19]. Dai et al. found significant distinctions in Weighted UniFrac between giant and small CBD stones [21]. Liu et al. reported significant differences in both Weighted and Unweighted UniFrac between choledocholithiasis cases and controls [28]. Biliary microbial composition in lithiasis In the current study, we reported alterations in microbial composition and the relative abundance of microbial taxa across different biliary lithiasis conditions (Table 2). Choledocholithiasis exhibited a significantly decreased relative abundance of Actinobacteria, Patescibacteria, Spirochaetota, Candidatus_Saccharibacteria, and Cyanobacteria compared to control groups [18]. Elevated levels of Bacteroidetes were observed in GSD by Molinero et al. [15]., while Proteobacteria was associated with CBD stones, particularly in those with recurrent CBD stones [20]. Choe et al. observed Actinobacteria and Firmicutes predominance in recurrent lithiasis and a higher frequency of Bacteroidetes in non-recurrent cases [19]. However, contrasting with these findings, Tan et al. found a decreased relative abundance of Actinobacteria alongside Bacteroidetes in non-recurrent cases [29]. Li and colleagues have identified core microbial orders such as Pasteurellales, Neisseriales, and Campylobacterales in cases of recurrent CBD stones. Conversely, distinct orders including Pseudomonadales, Burkholderiales, Bacillales, Aeromonadales, and Actinomycetales were noted in cases without recurrence [30]. At the family taxonomic level, individuals afflicted with GSD demonstrated elevated levels of Bacteroidaceae, Porphyromonadaceae, Prevotellaceae, Veillonellaceae, and compared to the control group, with the prevalence of Enterobacteriaceae being higher in the choledocholithiasis group compared to the GSD group [15, 23]. At the genus level, Bacteroides, Dialister, and Escherichia/Shigella were found to be more abundant in patients with GSD compared to controls [15]. Similarly, Zhitang Lyu et al. reported higher relative abundances of Brevundimonas and Prevotella, and lower abundances of various genera, including Actinomyces and Proteus, in patients with choledocholithiasis [24]. Park et al. highlighted that the genera Enterococcus, Lactobacillus, and Streptococcus were more abundant in GSD patients, whereas Prevotella, Bacteroides, and Ralstonia were more prevalent in CBD stones [23]. Importantly, the prevalence of Enterobacteriaceae was higher in the choledocholithiasis group compared to the GSD group [28, 29]. Discovering biliary candidate taxa as potential microbial indicators for differential diagnosis Bacterial analysis of potential microbial indicators is an effective approach for identifying unique microbial signatures between different groups. These represent hypothesis-generating findings rather than validated diagnostic markers, as no performance metrics (e.g., area under the curve (AUC), sensitivity, specificity) were reported in the included studies. Comparative studies on biliary lithiasis and non-stone-forming groups have revealed potential microbial indicators that highlight the pathophysiological distinctiveness of these conditions. Figures 3 and 4 illustrate the potential biliary microbial indicators found in various lithiasis conditions. Feng et al. identified Alcaligenaceae as a candidate microbial biomarker for biliary lithiasis with a linear discriminant analysis (LDA) score greater than 3 (LDA scores > 2 indicate the bacterial signature(s) with significant differential abundance between the compared groups), in contrast to a control group including patients with bile duct stenosis, PC, pancreatic mass, diverticulum, congenital bile duct abnormalities, and bile duct malignancy [22]. Similarly, Kim et al. found Enterococcus levels to be significantly higher (p = 0.031) in patients with choledocholithiasis compared to a diverse control group, with an LDA threshold of 2.0 [31]. Cai et al. further identified Prevotella, Klebsiella, Streptococcus anginosus, Morganella morganii, and Arthrobacter as possible microbial biomarkers for choledocholithiasis, while Bdellovibrionota, Rothia, Peptostreptococcus, Serratia, and Bifidobacterium bifidum were indicative of GSD when compared to liver donors without any HPB disorders, with LDA scores exceeding 2 [17]. Additionally, Lee et al. emphasized the Enterococcus genus as a potential biomarker for choledocholithiasis relative to a control group that included cases of dilated bile duct, benign biliary disorders, benign biliary stricture, CBD bile sludge, and autoimmune pancreatitis [25]. Chord diagram depicting phyla and orders identified as potential microbial indicators in various hepatopancreatobiliary disorders. Taxa were selected based on significant differential abundance reported in the included studies, primarily determined through LEfSe or similar methods. LDA thresholds were defined as scores > 2.0 (the default for indicating significant differential abundance), with some studies applying stricter criteria such as > 3.0 or > 4.0. The diagram shows taxa at the phylum and order levels of taxonomic resolution. Abbreviations: PC, pancreatic cancer; CCA, cholangiocarcinoma; CBD, common bile duct stone; RCBD, recurrent common bile duct stone; GSD, gallstone disease. Heatmap presenting potential microbial indicators in different hepatopancreatobiliary disorders compared to controls. The heatmap illustrates potential candidate taxa at the genus and species levels of taxonomic resolution, as reported in the included articles. Taxa were selected based on significant differential abundance identified in the reviewed studies, primarily using LEfSe or equivalent approaches. LDA thresholds were defined as scores > 2.0 (the standard cutoff for significance), though individual studies occasionally used higher thresholds like > 3.0 or > 4.0 for greater stringency. LDA scores ranging from − 6 to + 6 are displayed using a blue-to-red color gradient. Abbreviations: PC, pancreatic cancer; CCA, cholangiocarcinoma; CBD, common bile duct stone; RCBD, recurrent common bile duct stone; GSD, gallstone disease. Retrieved data on biliary bacterial biomarkers distinguishing between GSD and CBD stones also revealed unique microbial signatures. Feng et al. identified notable differences in bacterial families such as Propionibacteriaceae, Sphingomonadaceae, Lactobacillaceae, and Verrucomicrobiaceae (p = 0.001), along with the genus Campylobacter (p = 0.007) and species Propionibacterium acnes (p = 0.013) and Peptoclostridium difficile, as potential biomarkers specific to GSD compared to CBD stones [22]. Park et al. further identified Bacilli and Lactobacillales as possible microbial biomarkers with an LDA score exceeding 3 for GSD [23]. In reviewing the literature on recurrent stone formation and biliary microbial biomarkers, several studies have highlighted specific bacterial signatures linked to recurrent CBD stones. Tan et al. identified bacterial genera such as Kurthia, Paraclostridium, Enterobacter, Klebsiella, Raoultella, and Pyramidobacter as potential microbial indicators distinguishing recurrent CBD stones from non-recurrent cases [29]. Li et al. reported significant findings on bacterial taxa, identifying Fusobacteriales (p < 0.05) and Neisseriales (p < 0.05) as possible biomarkers specific to recurrent CBD stones [30]. Chen et al. elucidated families such as Enterococcaceae and Brevibacteriaceae, along with genera including Pyramidobacter, Enterococcus, Citrobacter, and Brevibacterium, as bacterial biomarkers with LDA scores exceeding 2.8 for recurrent CBD stones [26]. Liu’s research highlighted Enterococcus as a potential biomarker with an LDA score surpassing 4.5 for recurrent CBD stones [28]. In a specialized investigation into different types of large and small CBD stones, specific bacterial indicators were identified for each size category. The study revealed that Enterococcus, Citrobacter, Lactobacillus, Pyramidobacter, Bifidobacterium, and Shewanella were significantly associated with the formation of large stones, whereas Granulicatella and Fusobacterium were markers for small stone formation. These associations were supported by LDA scores exceeding 2 [21]. Biliary microbiome in patients with PC Despite mounting evidence suggesting a link between PC and microbial communities in the oral cavity, blood, duodenal fluid, gut, and tumor microenvironment [32], data on the role of biliary microbiome in this malignancy remain scarce. To address this gap, we reviewed three studies encompassing 45 PC patients and 45 control individuals, including those with GSD, CBD stones, and CCA. Biliary bacterial diversity in PC Understanding the diversity of bacterial taxa in PC could provide valuable insights into the microbial factors involved in the development of this disease. However, there is limited data regarding changes in biliary bacterial diversity associated with PC. These findings highlight the distinct microbial diversity patterns in PC, emphasizing the need for further research to elucidate the role of the biliary microbiome in PC development. Bacterial alpha diversity in PC In a study by Li et al., the Chao1 index was significantly higher in patients with PC compared to those with GSD (p < 0.05) [27]. Poudel et al. reported no significant differences in Shannon, Simpson, Fisher alpha, or Pielou evenness between PC and CCA [33]. Merali et al. found no significant differences in Shannon or Simpson between PC and benign biliary diseases [34] (Fig. 2A). Bacterial beta diversity in PC Merali et al. reported a significant difference in Aitchison distance between PC and benign biliary diseases (p = 0.0173) [34] (Fig. 2B). Biliary microbial composition in PC The relative abundance of biliary bacteria in PC compared to non-PC groups has been the focus of a few key studies (Table 3). Poudel et al. found that the genus Rothia was significantly more prevalent in PC patients than in those with CCA (p = 0.008) [33]. Additionally, Merali et al. reported that the genus Streptococcus was more frequently detected in patients with PC compared to those with benign biliary diseases (p = 0.033) [34]. These findings suggest that certain bacterial genera may be associated with the presence of PC, highlighting the potential role of the biliary microbiome in the pathogenesis of this malignancy. The observed differences in bacterial abundance between PC and non-PC groups underscore the importance of further research to explore these microbial associations and their potential implications for early detection, diagnosis, and treatment of PC. Table 3Characteristics of the biliary Microbiome alternations associated with pancreatic cancer: microbial abundance, diversity, and biomarkers across reviewed studiesPancreatic cancerArticleCase/ControlEnriched AbundanceDepleted AbundancePotential BiomarkerAlpha DiversityBeta DiversityLi et al. [27]PC/Cholelithiasis--↑= g-Pseudomonas, g-Chloroplast, g-Acinetobacter, g-Allorhizobium_g-Neorhizobium_g-Pararhizobium_g-Rhizobium_ g-Rhizobium, g-Exiguobacterium, g-Halomonas, g-Staphylococcus, g-Cupriavidus, g-Bacillus, g-Cutibacterium, g-Cloacibacterium, f-Vicinamibacteraceae, g-Shewanella,, g-Geobacter, g-Corynebacterium,, g-Brevundimonas,, g-Rubellimicrobium, g-Nitrospira, g-Massilia, g-Rokubacteriales, g-Gaiella,↓= g-Faecalibaculum, g-ParasutterellaChao1: Sig higher (p < 0.05)-Poudel et aal. [33]PC/CCAg-Rothia (p = 0.008)g-Akkermansia (p = 0.031), g-Achromobacter (p = 0.031)-Shannon: NSSimpson: NSFisher alpha: NSPielou evenness: NS-Merali et al. [34]PDAC/Benign biliary diseasesg-Streptococcus (FDR = 0.033)g-Escherichia (FDR = 0.002), g-Unclassified genus from proteobacteria (FDR = 0.002), g-Unclassified genus from Enterobacteriaceae (FDR = 0.011)-Shannon: NSSimpson: NSAitchison distance: Sig (p = 0.0173)PC pancreatic cancer, CCA cholangiocarcinoma, PDAC pancreatic ductal adenocarcinoma, FDR false discovery rate, NS not-significant, Sig significantThe symbols “↓” and “↑” indicate a decrease and an increase in diversity, respectively Characteristics of the biliary Microbiome alternations associated with pancreatic cancer: microbial abundance, diversity, and biomarkers across reviewed studies ↑= g-Pseudomonas, g-Chloroplast, g-Acinetobacter, g-Allorhizobium_g-Neorhizobium_g-Pararhizobium_g-Rhizobium_ g-Rhizobium, g-Exiguobacterium, g-Halomonas, g-Staphylococcus, g-Cupriavidus, g-Bacillus, g-Cutibacterium, g-Cloacibacterium, f-Vicinamibacteraceae, g-Shewanella,, g-Geobacter, g-Corynebacterium,, g-Brevundimonas,, g-Rubellimicrobium, g-Nitrospira, g-Massilia, g-Rokubacteriales, g-Gaiella, ↓= g-Faecalibaculum, g-Parasutterella Shannon: NS Simpson: NS Fisher alpha: NS Pielou evenness: NS Shannon: NS Simpson: NS PC pancreatic cancer, CCA cholangiocarcinoma, PDAC pancreatic ductal adenocarcinoma, FDR false discovery rate, NS not-significant, Sig significant The symbols “↓” and “↑” indicate a decrease and an increase in diversity, respectively Identifying biliary microbial as a potential microbial indicator for early detection of PC Microbiome-based potential indicators play a crucial role in cancer research, especially for PC. Figures 3 and 4 present potential microbial biomarkers identified in the bile of patients with PC across various taxonomic levels. In this context, Li et al.. performed a linear discriminant analysis effect size (LEfSe) analysis to improve the diagnostic accuracy of PC. Their study identified specific bacterial taxa in the bile that could serve as candidate taxa for PC, including Pseudomonas, Chloroplast, and Acinetobacter. These taxa were significantly associated with PC, suggesting their potential utility in distinguishing PC from GSD conditions. Conversely, the study also identified Faecalibaculum and Parasutterella as potential biomarkers for cholelithiasis compared to pancreatic cancer, with an LDA score exceeding 3 [27]. This differential presence of bacterial taxa highlights the distinct microbial profiles associated with different pancreaticobiliary diseases and underscores the importance of further research to validate these findings and explore their potential clinical applications in the diagnosis and management of pancreatic cancer. Biliary microbiome in patients with CCA Dysbiosis in the biliary microbiome may contribute to CCA development by promoting inflammation and facilitating genetic and epigenetic alterations, especially in conditions such as CCA [35]. Of seven studies on CCA, there were 129 CCA patients and a control cohort of 206 predominantly lithiasis-afflicted individuals. Our primary objective was to elucidate microbial diversity, relative abundance patterns, and potential biomarkers involved in CCA pathogenesis. Biliary bacterial diversity in CCA As shown in Table 4, these findings highlight the distinct microbial profiles associated with CCA, suggesting the potential of microbial diversity as a diagnostic tool for differentiating CCA from other biliary conditions. Table 4Characteristics of the biliary Microbiome alterations associated with cholangiocarcinoma: microbial abundance, diversity, and biomarkers across reviewed studiescholangiocarcinomaArticleCase/ControlEnriched AbundanceDepleted AbundancePotential BiomarkerAlpha DiversityBeta DiversityDangtakot et al. [36]iCCA/Choledocholithiasisp-proteobacteria, p-cyanobacteria, p-acidobacteria, p-chlorobi, p-fibrobacteres, p-armatimonadetes, p-gemmatimonadetes, g- g-Enterobacter, g-pseudomonas, s-Enterobacter (MT975285), s-enterococcus faecalis, s-Pseudomonas aeruginosa, s-Stenotrophomonas geniculatep-Fusobacteria, p-synergistetes, g-Cetobacterium g-pyramidobacter, g-streptococcus, s-Cetobacterium (MH755457), s-Bacteroides pyogenes, s-Prevotella heparinolytica, s-Fusobacterium periodonticum, s-Clostridium perfringens, s-Bacteroides Vulgatus↑= g-Enterobacter,o-pseudomonadales, g-pseudomonas,f-pseudomonadaceace, s-pseudomonas aeruginosa↓= c-Fusobacteria, o-fusobacteriales, p-fusobacteria, f-fusobacteriacease, g-Cetabacterium, g-bacteroides, f-Bacteroidaceae, f-Synergistaceae, g-pyramidobacter, o-Synergistales, c-synergistia, p-synergistetes, g-edwardsiella, o-clostridiales, c-clostridiaObserved OTU: Sig higherChao1: Sig higherACE: Sig higherShannon: Sig lowerSimpson: Sig lower-Jimenez et al. [38]eCCA/Benign biliary diseasesP-Fusobacteria, p-Acidobacteria, p-Planctomycetes (p < 0.0.05)//f-Methylophilaceae, f-Sinobacteriaceae, g-Actinomyces, g-Dialister, g-Novosphingobium, g-Prevotella, g-Fusobacterium, g-H.pylori, g-streptococcuss = Prevotella melaninogenicao-rhizobialesf-micrococcaceace, f-phyllobacteriaceaeg-Nesterenkonia, g-Mesorhizobiums-rothia mucilaginosa--Unweighted UniFrac: Sig (p = 0.01)Saab et al. [39]eCCA/Cholelithiasisp-Deinococcus-Thermus, p-Gemmatimonadetesg-Bacteroides, g-Geobacillus, g-Meiothermus, g-Anoxybacillusp-Firmicutes, p-Fusobacteria, p-Actinobacteria, p-Tenericutesg-Klebsiella, g-Clostridium, g-Fusobacterium, g-Haemophilus, g-Enterobacter, g-Citrobacter, g-Actinomyces--Bray-Curtis: Sig (p = 0.01)Chen et al. [26]dCCA/Choledocholithiasisp-Chloroflexi (p = 0.000), p-Gemmatimonadetes(p = 0.000), p-Nitrospirae(p = 0.000), p-Latescibacteria (p = 0.021), p-Planctomycetes (p = 0.039), g- Staphylococcus (p = 0.000), g-Okibacterium(p = 0.000), g-Corynebacterium(p = 0.000), g-Sphingomonas (p = 0.001),g-Stenotrophomonas(p = 0.002),g-Acidovorax(p = 0.002),Gemmiger(p = 0.002),Providencia(p = 0.002),g-Comamonas(p = 0.002),g-Methylobacterium(p = 0.002)--Observed OTU: NSChao1: NSShannon: Sig higher (p = 0.03)Simpson: NSPoudel et al. [33]CCA/PCg-Akkermansia (p = 0.031), g-Achromobacter (p = 0.031)g-Rothia-Shannon: NSSimpson: NSFisher alpha: NSPielou evenness: NS-Li et al. [27]pCCA/Cholelithiasis--↑= g-Pseudomonas, g-Sphingomonas, g-Halomonas, g-Prevotella, g-Shewanella, g-Bacillus, g-Corynebacterium, g-Chloroplast, g-Lactococcus, Ellin516, g-Catenisphaera, f_67_14, andg-Catenibacterium↓=g-Escherichia-Shigella, g-Neisseria, g-Stenotrophomonas, g-Turicibacter--Li et al. [27]dCCA/Cholelithiasis--↑= g-Streptococcus, g-Prevotella, g-Halomonas, g-Helicobacter, g-Rikenellaceae_RC9_gut_group, g-Actinobacillus, g-Lactococcus, g-Pseudomonas, g-Catenisphaera, g-Devosia, g-Mobiluncus, g-Sphingobacterium, g-Comamonas↓= g-Escherichia-ShigellaShannon: Sig higherSimpson: Sig higher-Miyabe et al. [37]CCA/Biliary lithiasisp-Firmicutes, p-Fusobacteria, and p-Actinobacteria-Observed OTU: Sig higher-CCA cholangiocarcinoma, iCCA intrahepatic cholangiocarcinoma, eCCA extrahepatic cholangiocarcinoma, dCCa distal cholangiocarcinoma, pCCA perihilarcholangiocarcinoma, OUT observational taxonomic unit, NS not-significant, Sig significantThe symbols “↓” and “↑” indicate a decrease and an increase in diversity, respectively Characteristics of the biliary Microbiome alterations associated with cholangiocarcinoma: microbial abundance, diversity, and biomarkers across reviewed studies ↑= g-Enterobacter,o-pseudomonadales, g-pseudomonas,f-pseudomonadaceace, s-pseudomonas aeruginosa ↓= c-Fusobacteria, o-fusobacteriales, p-fusobacteria, f-fusobacteriacease, g-Cetabacterium, g-bacteroides, f-Bacteroidaceae, f-Synergistaceae, g-pyramidobacter, o-Synergistales, c-synergistia, p-synergistetes, g-edwardsiella, o-clostridiales, c-clostridia Observed OTU: Sig higher Chao1: Sig higher ACE: Sig higher Shannon: Sig lower Simpson: Sig lower P-Fusobacteria, p-Acidobacteria, p-Planctomycetes (p < 0.0.05)//f-Methylophilaceae, f-Sinobacteriaceae, g-Actinomyces, g-Dialister, g-Novosphingobium, g-Prevotella, g-Fusobacterium, g-H.pylori, g-streptococcus s = Prevotella melaninogenica o-rhizobiales f-micrococcaceace, f-phyllobacteriaceae g-Nesterenkonia, g-Mesorhizobium s-rothia mucilaginosa p-Deinococcus-Thermus, p-Gemmatimonadetes g-Bacteroides, g-Geobacillus, g-Meiothermus, g-Anoxybacillus p-Firmicutes, p-Fusobacteria, p-Actinobacteria, p-Tenericutes g-Klebsiella, g-Clostridium, g-Fusobacterium, g-Haemophilus, g-Enterobacter, g-Citrobacter, g-Actinomyces Observed OTU: NS Chao1: NS Shannon: Sig higher (p = 0.03) Simpson: NS Shannon: NS Simpson: NS Fisher alpha: NS Pielou evenness: NS ↑= g-Pseudomonas, g-Sphingomonas, g-Halomonas, g-Prevotella, g-Shewanella, g-Bacillus, g-Corynebacterium, g-Chloroplast, g-Lactococcus, Ellin516, g-Catenisphaera, f_67_14, and g-Catenibacterium ↓=g-Escherichia-Shigella, g-Neisseria, g-Stenotrophomonas, g-Turicibacter ↑= g-Streptococcus, g-Prevotella, g-Halomonas, g-Helicobacter, g-Rikenellaceae_RC9_gut_group, g-Actinobacillus, g-Lactococcus, g-Pseudomonas, g-Catenisphaera, g-Devosia, g-Mobiluncus, g-Sphingobacterium, g-Comamonas ↓= g-Escherichia-Shigella Shannon: Sig higher Simpson: Sig higher CCA cholangiocarcinoma, iCCA intrahepatic cholangiocarcinoma, eCCA extrahepatic cholangiocarcinoma, dCCa distal cholangiocarcinoma, pCCA perihilarcholangiocarcinoma, OUT observational taxonomic unit, NS not-significant, Sig significant The symbols “↓” and “↑” indicate a decrease and an increase in diversity, respectively Bacterial alpha diversity in CCA Significantly elevated Observed OTU, Chao1, and ACE indices have been reported in CCA patients compared to those with CBD stones [36]. Miyabe highlighted a substantial increase in observed OTU in CCA patients compared to the lithiasis group [37]. Chen et al. observed a markedly higher Shannon index (p = 0.03) in CCA patients compared to those with choledocholithiasis [26]. Li et al. found significantly elevated trends in Shannon and Simpson indices among CCA patients compared to GSD patients [27]. Conversely, Chen et al. reported no significant differences in Observed OTU, Chao1, or Simpson between dCCA and choledocholithiasis [26]. Poudel et al. found no significant differences in Shannon, Simpson, Fisher alpha, or Pielou evenness between CCA and PC [33]. Bacterial beta diversity in CCA Significant distinctions were found in the Unweighted UniFrac index (p = 0.01) between CCA patients and control groups [38]. Similarly, Saab’s study indicated significant distinctions in the Bray-Curtis index (p = 0.01) between CCA patients and individuals with GSD [39]. Biliary bacterial relative abundance in CCA Significant differences in bacterial taxa between CCA patients and control groups were observed in various studies (Table 4). Dangtakok et al. noted higher relative abundances of the phyla Proteobacteria, Cyanobacteria, Gemmatimonadetes, Acidobacteria, Chlorobi, Fibrobacteres, and Armatimonadetes, as well as the genera Enterobacter, Pseudomonas, and species such as Enterobacter (MT975285), Enterococcus faecalis, Pseudomonas aeruginosa, and Stenotrophomonas geniculate in CCA patients compared to those with CBD stones [36]. Avilés-Jiménez et al. highlighted the increased prevalence of the phyla Fusobacteria, Acidobacteria, and Planctomycetes; the families Methylophilaceae and Sinobacteriaceae; and the genera Actinomyces, Dialister, Novosphingobium, Prevotella, Fusobacterium, and Streptococcus. They also noted an increased prevalence of the species Prevotella melaninogenica and Helicobacter pylori in patients with CCA compared to controls [38]. Saab’s study reported higher relative abundances of the phyla Deinococcus-Thermus and Gemmatimonadetes, and the genera Bacteroides, Geobacillus, Meiothermus, and Anoxybacillus in CCA patients compared to those with GSD [39]. Chen et al. identified the phyla Chloroflexi, Gemmatimonadetes, Nitrospirae, Latescibacteria, and Planctomycetes, as well as the genera Staphylococcus, Okibacterium, Corynebacterium, Sphingomonas, Stenotrophomonas, Acidovorax, Gemmiger, Providencia, Comamonas, and Methylobacterium as more prevalent in CCA patients compared to those with CBD stones [26]. Poudel et al. highlighted significantly higher levels of the genera Akkermansia and Achromobacter in CCA patients compared to those with PC [33]. Finally, Miyabe et al. demonstrated a greater prevalence of the phyla Firmicutes, Fusobacteria, and Actinobacteria in CCA patients compared to those with biliary lithiasis [37]. Biliary candidate taxa as potential microbial indicators in CCA Identifying bacterial biomarkers is a novel and interesting approach for distinguishing CCA from other biliary diseases. A recent study has identified several microbial taxa as potential biomarkers for distinguishing CCA patients from those with CBD stones (Fig. 3). Notable taxa include Enterobacter, Pseudomonadales, Pseudomonas, Pseudomonadaceae, and P. aeruginosa, all with LDA scores exceeding 3.6 [36]. In a comprehensive study, Li et al. recognized specific microbial taxa as potential biomarkers for perihilar cholangiocarcinoma (pCCA) and distal cholangiocarcinoma (dCCA). For pCCA, Pseudomonas, Sphingomonas, Halomonas, Prevotella, and Shewanella were proposed as biliary candidate taxa. For dCCA, Streptococcus, Prevotella, and Halomonas were identified as distinguishing markers when compared to GSD, all with LDA scores exceeding 3 [27]. These findings highlight the significant potential of microbial markers in enhancing the diagnostic accuracy of CCA and distinguishing it from biliary stones. Biliary microbiome in patients with PSC PSC is characterized by chronic inflammation and damage to both intra- and extrahepatic bile ducts. Numerous studies have demonstrated that patients with PSC present an altered microbiome within the bile, indicating a potential association between microbial populations and the development of the disease [40, 41]. Specifically, in PSC, individuals frequently exhibit significant biliary dysbiosis compared to their controls. This microbial shift has led some researchers to speculate about its potential pathogenic role in triggering the inflammatory responses underlying PSC [40]. In our study, we selected two studies that included a total of 123 patients with PSC and 46 non-PSC diagnostic controls, including individuals with choledocholithiasis and papillary adenoma. Biliary bacterial diversity in PSC Alterations in bile microbial diversity have been observed in PSC patients compared to non-PSC individuals (Table 5). Table 5Characteristics of the biliary Microbiome alternations associated with primary sclerosing cholangitis: microbial abundance, diversity, and biomarkers across reviewed studiesPrimary sclerosing cholangitisArticleCase/ControlEnriched AbundanceDepleted AbundancePotential BiomarkerAlpha DiversityBeta DiversityLiwinski et al. [42]PSC/CholedocholithiasisP-proteobacteria (p = 0.038), g-Enterococcus (FDR = 0.000), g-Staphylococcus (FDR = 0.000), g-Neisseria (FDR = 0.000), g-enhydrobacter, g-prevotella, g-lawsonella, g-sphingomonas, g-cutibacterium, s-Enterococcus faecalis, s-Staphylococcus epidermidis, s-Streptococcus sanguinis, s-Enhydrobacter aerosaccus, s-Prevotella pallens, s-Veillonella dispar--Shannon: Sig lower (p = 0.012)Bray-Curtis: Sig (p < 0.01)Pereira et al. [43]Early PSC/Controlf-Pasteurellaceae, f-Staphylococcaceaeg-Neisseria (Otu00045), g-Campylobacter (Otu00089),f-Enterobacteriaceae (Otu00 008), f-Neisseriaceae (Otu00 213), f-Xanthomonadaceae-Shannon: NS-Pereira et al. [43]Advanced PSC/Early PSCg-Streptococcus--Shannon: NS-PSC primary sclerosing cholangitis, NS Not-significant, Sig significant Characteristics of the biliary Microbiome alternations associated with primary sclerosing cholangitis: microbial abundance, diversity, and biomarkers across reviewed studies f-Pasteurellaceae, f-Staphylococcaceae g-Neisseria (Otu00045), g-Campylobacter (Otu00089), PSC primary sclerosing cholangitis, NS Not-significant, Sig significant Bacterial alpha diversity in PSC Liwinski et al. reported a significantly lower Shannon index in the PSC group compared to controls with choledocholithiasis and papillary adenoma (p = 0.012) [42]. Pereira et al. found no significant differences in Shannon between early PSC and controls or advanced versus early PSC [43]. Bacterial beta diversity in PSC Beta diversity analysis using the Bray-Curtis index showed significant differences between the PSC and control groups (p ≤ 0.01) [42]. Biliary bacterial composition in PSC Several studies have explored the altered microbiome in PSC patients (Table 5). Liwinski et al. reported that specific taxa, including the phylum Proteobacteria (p = 0.038) and the genera Enterococcus (p = 0.000), Staphylococcus (p = 0.000), Neisseria (p = 0.000), Enhydrobacter, Prevotella, Lawsonella, Sphingomonas, and Cutibacterium, were significantly more prevalent in PSC patients compared to controls, including choledocholithiasis and papillary adenoma. Specific species such as Enterococcus faecalis, Staphylococcus epidermidis, Streptococcus sanguinis, Enhydrobacter aerosaccus, Prevotella pallens, and Veillonella dispar were also more common in PSC patients [42]. Furthermore, Pereira et al. suggested that the families Pasteurellaceae and Staphylococcaceae, as well as the genera Neisseria (OTU00045) and Campylobacter (OTU00089), were more frequently observed in PSC patients compared to non-PSC control groups [43]. These findings underscore the significance of microbial markers in understanding the pathogenesis of PSC and highlight potential targets for diagnostic and therapeutic interventions. The evolving microbiome landscape in PSC presents a compelling avenue for research; however, the precise contribution of these microbial alterations to the pathogenesis of the condition remains largely unknown. Discussion A growing body of research underscores the pivotal role of a dysbiotic biliary microbiome in human diseases, particularly HPB disorders [44]. This systematic review aims to summarize and synthesize evidence from previous studies examining alterations in microbiome composition within the bile tract and their potential roles in the development and progression of HPB diseases. Our attempt at bile microbiome changes in HPB diseases offers important insights into various types of chronic and malignant HPB diseases. Our review highlights significant variations in bacterial diversity across different HPB conditions, with studies showing mixed results regarding microbial diversity indices. Some studies have reported reduced alpha diversity in the biliary microbiome of patients with malignant conditions compared to controls [36, 42]. This reduction in microbial diversity might be associated with disease progression and the inflammatory environment it creates. Conversely, other studies have observed higher alpha diversity in diseased groups, suggesting that microbial diversity might increase in response to the tumor microenvironment [26, 27, 36, 37]. Despite the contradicting findings on alpha diversity, there is a consistent observation of distinct microbial communities (beta diversity variances) in patients with malignant biliary conditions compared to those with benign conditions [34, 38, 39]. These differences in beta diversity indicate that specific microbial profiles may be associated with pancreaticobiliary malignancies. To conclude, the biliary microbiome exhibits notable diversity change in malignant conditions compared to benign ones. Our review identifies specific bacteria associated with various HPB diseases, highlighting pathogenic mechanisms and candidate taxa as potential microbial indicators. For instance, the phylum Planctomycetes was reported to be enriched in the bile of CCA patients [26, 38] and has also been detected in other cancer types, such as in blood samples recruited from leukemia patients [45]. Proteobacteria were prevalent in CCA, PSC, and recurrent stone formation, with studies suggesting their significant role in these diseases due to their potential to promote chronic inflammation [20, 29, 36, 42, 46, 47]. The Pseudomonas genus is associated with both CCA and PC, with findings indicating its potential pathologic effects in tumorigenesis [27, 36, 48, 49]. The Staphylococcus genus is linked to CCA, PSC, and PC, with higher abundance noted in these conditions [26, 27, 42, 50]. The Klebsiella genus is significant in stone diseases, containing enzymes that facilitate biliary stone formation [17, 28, 29]. Additionally, biliary Klebsiella pneumoniae has been implicated in chemotherapy failure in PC patients [51]. In lithiasis conditions, Pyramidobacter has been reported to be more prevalent in recurrent CBD stones, and Citrobacter, with its antimicrobial resistance and virulence factors, is also prevalent in both recurrent and giant CBD stones [21, 26, 29]. These findings highlight the importance of specific bacterial taxa in the pathogenesis of HPB diseases, which is why further investigations should be carried out to uncover the underlying roles of potential microbial indicators in HPB diseases. Among studies on cholelithiasis, the prognostic significance of the biliary microbiome has been of great importance. The consistent association of specific bacterial genera with stone size and recurrence highlights the potential of microbial profiling as a valuable prognostic tool [21, 26, 28–30]. Additionally, various studies have investigated the association between cholelithiasis and the development of HPB malignancies, demonstrating that the formation of stones could be a risk factor and an early stage of carcinogenesis in the future [26, 27, 36, 37, 39]. The identified similarities in biliary microbial profiles may provide insights into the underlying pathogenesis of primary inflammatory conditions in the bile ducts and their potential progression to related malignancies. On the other hand, the differences in biliary microbial signatures reported across various conditions could indicate prognostic factors associated with altered microbial communities in the bile. Hence, these findings may have significant potential as screening tools for assessing the risk of developing HPB cancers from benign HPB conditions. However, despite its potential to distinguish between HPB disorders and controls, using the biliary microbiome as a diagnostic tool is invasive due to the necessity of an ERCP procedure or surgery for sampling. Additionally, the significant enrichment of environmental opportunistic microorganisms, such as Enterococcus and Klebsiella, may partly result from device-associated contamination during endoscopic procedures [52, 53]. To prevent contamination and misinterpretation of the results, numerous measures were considered during sample collection, experimental procedures, and data interpretation, which are worth mentioning. These can be categorized into procedural safeguards, reagent controls, and analytical filtering to systematically mitigate device-associated contamination, particularly during ERCP or similar interventions where endoscopes may introduce environmental or duodenal microbes. For instance, Avilés-Jiménez et al. found 5 OTU from Comamonadaceae and Pseudomonadales in reagent controls and excluded them, using water blanks to identify contamination [38]. In bile sampling, researchers using ERCP employed strict methods to prevent the introduction of microbes from the mouth or duodenum. Saab et al. and Merali et al. collected bile directly with a catheter before any contrast or brushing [34, 39], while Kim et al. and Lee et al. waited 24 h and collected bile through endoscopic nasobiliary drainage (ENBD) tubes to minimize contamination from the upper gut [25, 31]. Lyu et al. and Ye et al. emphasized the importance of sterile technique, including flushing the endoscope channel with water or saline and avoiding suction [20, 24]. Additionally, some studies incorporated negative controls (e.g., sterile water or blank extractions) during sequencing to identify and subtract potential contaminants like Pseudomonas or other environmental taxa [42]. Analytically, tools like QIIME’s decontamination filters or manual OTU exclusion based on blank samples were used in studies by Molinero et al. and others to refine datasets [15]. Despite these efforts, achieving a completely contamination-free state remains challenging, as residual device-related biases, such as the enrichment of opportunistic taxa like Enterococcus or Klebsiella from endoscope surfaces, may still affect results [54]. This could obscure disease-specific signatures, underscoring the need for standardized protocols in future research. It is noteworthy to mention that the microbiome not only can adhere to host cells but can also modulate metabolic pathways, which may contribute to pathogenic conditions such as stone formation and cancer development [55]. For instance, in gallstone disease, taxa like P. aeruginosa have been shown to secrete β-glucuronidase and phospholipase A2, enzymes that promote cholesterol crystallization and stone formation by shortening nucleation time [44]. In cholestatic conditions such as PSC and primary biliary cholangitis (PBC), dysbiosis leads to reduced conversion of primary to secondary bile acids, exacerbating inflammation and fibrosis through impaired farnesoid X receptor (FXR) signaling and increased toxic bile acid accumulation [56]. Key taxa like Clostridium, Bifidobacterium, and Lactobacillus contribute to these processes via bile salt hydrolase (BSH) activity, providing energy sources and enhancing bile resistance, while reduced butyrate-producers (e.g., Faecalibacterium) impair intestinal barrier function [57]. Although functional predictions like Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) were underutilized in the included studies, with only a few inferring metabolic pathways (e.g., tryptophan and ceramide metabolism in cholangitis [5 from Springer article]), future research should incorporate advanced tools such as HUMAnN for metagenomic pathway analysis to better elucidate the metabolic roles of these taxa in HPB disorders. The reviewed studies predominantly relied on LEfSe and associated LDA scores to identify differentially abundant taxa as potential microbial indicators. While LEfSe is a widely used tool for biomarker discovery in microbiome research due to its ability to incorporate statistical significance and effect size, it has notable limitations, particularly in addressing the compositional nature of microbiome data. For instance, LEfSe assumes independence among features and may produce inflated false positives in sparse datasets or when compositional constraints (e.g., relative abundances summing to 1) are not adequately accounted for, potentially leading to overestimation of differential taxa [58]. Some included studies mitigated this by incorporating complementary approaches, such as Aitchison distance for beta diversity (e.g., in pancreatic cancer analyses), which better handles compositional data through centered log-ratio transformations. To enhance robustness and reproducibility, future studies should consider integrating methods like Analysis of Compositions of Microbiomes with Bias Correction (ANCOM-BC) or other compositionally aware techniques, which adjust for biases in relative abundance data and provide more reliable differential abundance estimates. It is important to distinguish between hypothesis-generating biomarkers and validated diagnostic markers in the context of microbiome research. Hypothesis-generating biomarkers, such as the candidate taxa identified in this review through exploratory methods like LEfSe, represent preliminary microbial signatures associated with HPB disorders based on differential abundance analyses. These are derived from observational data and serve to generate testable hypotheses about potential roles in pathogenesis or diagnostics, but do not imply clinical utility. In contrast, validated diagnostic markers require rigorous evaluation in independent, prospective cohorts, including assessment of performance metrics such as AUC, sensitivity, and specificity to demonstrate reliability, reproducibility, and clinical relevance. Notably, none of the included studies reported such metrics for the identified taxa, as their focus was on hypothesis generation rather than diagnostic validation. This underscores the exploratory nature of the current evidence and highlights the need for future studies to develop and validate microbiome-based models with quantitative performance evaluations before considering them for diagnostic applications. The study of the biliary microbiome is constrained by several methodological and interpretive limitations. A primary challenge is the inherent difficulty in sample acquisition; bile collection is an invasive procedure, typically restricted to patients undergoing ERCP, liver transplantation, or other surgeries. This clinical context introduces significant confounding variables, as samples are often obtained from individuals who have received antibiotic treatments, potentially altering the native microbial composition. Furthermore, the naturally low bacterial biomass in bile complicates precise genomic analysis and increases the susceptibility to contamination, particularly during procedures like ERCP, thereby introducing potential biases in microbial profiling. These collection constraints have direct consequences for research design and generalizability. The limited specimen availability has impeded the establishment of a definitive healthy biliary microbiome baseline. Consequently, many studies lack appropriate healthy control groups and instead rely on heterogeneous comparators, such as patients with benign hepatobiliary conditions. This heterogeneity poses a major challenge for cross-study comparisons. The scarcity of samples also results in a limited number of studies for specific disease categories, such as PSC, which hinders the identification of consistent, disease-associated microbial taxa. Compounding these issues is a lack of methodological standardization across the field. The use of inconsistent sequencing regions and bioinformatic pipelines obstructs the clarification of the microbiota’s role in HPB diseases and prevents robust meta-analyses. To overcome these challenges and elucidate the potential of biliary microbial signatures as biomarkers for disease progression, future research must transition from case-control designs to prospective longitudinal cohorts. Such an approach would enable the critical assessment of microbial dynamics over time and their relationship to clinical outcomes. In summary, these findings highlight the importance of conducting future studies with larger sample sizes and more rigorous methodologies; however, despite their promise, the diagnostic utility of the biliary microbiome remains speculative given the heterogeneity in methodologies and the small number of studies with healthy controls. As such, further prospective validation is required before translation into clinical practice, including evaluation of performance metrics such as AUC, sensitivity, and specificity for any candidate taxa identified as potential diagnostic markers. Such studies should aim to include control groups without any HPB disorders and consider the effects of antibiotic treatments on microbial composition. Additionally, utilizing advanced bioinformatics tools and novel pipelines will enhance our ability to analyze low-abundance microbial communities in the bile. Understanding the biliary microbiome’s role in HPB diseases could lead to the development of effective diagnostic and therapeutic strategies, ultimately improving patient outcomes. By compiling all candidate taxa as potential biomarkers related to HPB conditions in bile specimens, this systematic review can pave the way for further research on the role of the biliary microbiome in disease pathogenesis. This could help prevent the progression from chronic inflammatory conditions, such as PSC and cholelithiasis, to cancerous conditions. Furthermore, it highlights the importance of ongoing research to develop non-invasive tests for early detection based on potential microbiome biomarkers. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements We are grateful to the members of the Research Institute for Gastroenterology and Liver Diseases, Shahid Beheshti University of Medical Sciences, Tehran, Iran. Abbreviations Analysis of Compositions of Microbiomes with Bias Correction Amplicon sequence variants Area under the curve Bile salt hydrolase Common bile duct Cholangiocarcinoma Computed tomography Distal cholangiocarcinoma Extrahepatic cholangiocarcinoma European Nucleotide Archive Endoscopic nasobiliary drainage Endoscopic retrograde cholangiopancreatography Endoscopic ultrasound Farnesoid X receptor Global Burden of Disease Gallstone disease Hepatopancreatobiliary Intrahepatic cholangiocarcinoma Intensive care unit Linear discriminant analysis Linear discriminant analysis effect size Magnetic resonance cholangiopancreatography Magnetic resonance imaging Non-alcoholic steatohepatitis Newcastle–Ottawa Scale Operational taxonomic units Primary biliary cholangitis Pancreatic cancer Perihilar cholangiocarcinoma Pancreatic ductal adenocarcinoma Phylogenetic Investigation of Communities by Reconstruction of Unobserved States Preferred Reporting Items for Systematic Reviews and Meta-Analyses Primary sclerosing cholangitis Percutaneous transhepatic biliary drainage World Health Organization Authors’ contributions HR.H., Z.H., and M.R.N. conceptualized the review. H.H., Z.H., and N.A. conducted the systematic review, collated and analyzed the data, and drafted the manuscript. B.K. contributed to the content of the figures. HR.H. supervised the project, reviewed the manuscript, and provided final approval. Funding The study was supported by the Research Institute for Gastroenterology and Liver Diseases, Shahid Beheshti University of Medical Sciences, Tehran, Iran (NO. IR.SBMU.RIGLD.REC.1404.036). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Hossein Halimi and Zeinab Hesami contributed equally to this work. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement No datasets were generated or analysed during the current study.