Liquid biopsy (LB) represents the sampling and analysis of human body fluids including blood, urine, saliva, cerebrospinal or pleural fluid, ascites, tears, sweat, and bile as a minimally invasive method for the diagnosis and prognosis of different diseases [1]. The rationale implies that biofluids are loaded with a variety of analytes (DNA, proteins, metabolites, extracellular vesicles, or cells) where the detection of specific biomarkers can distinguish between health and disease or even inform about disease progression and response to treatment. Accordingly, the number of clinical trials for LB in different therapeutic areas is growing dramatically, with more than 1,500 trials using blood in the oncology field [2] for early disease detection, to guide therapy, and to monitor outcome and residual disease [2]. However, sensitivity and specificity represent key challenges in the development of LB strategies for cancer screening, and the selection of more reliable biomarkers and/or the development of sample enrichment procedures represent areas of intense research [1, 3]. In addition, the use of biofluids other than blood could overcome some limitations, including blood biomarker concentration and specificity avoiding for instance clonal hematopoiesis mutations [4].

In this sense, as recently reviewed [5–10], bile has emerged as a promising LB matrix for the management of patients with pancreaticobiliary tumors causing biliary strictures. Here, the most recent findings on the utility of bile circulating cell-free DNA (cfDNA) analyses in those oncological conditions are reviewed. Moreover, different clinical situations where bile could be available, the current bile collection procedures, and the potential clinical applications of bile analyses are presented.

Bile is constantly produced by hepatocytes, stored in the gallbladder, and secreted into the duodenum through the biliary tract. Bile is a complex fluid composed of bile acids, phospholipids, cholesterol, bilirubin, proteins, inorganic salts, extracellular vesicles, DNA, and RNA [11–14]. Bile composition can change in different pathological conditions; therefore, the identification of these changes can be harnessed into good diagnostic tools, can inform about pathological processes, or can even help in directing treatments [5, 9, 14–16]. Useful analytes include small molecules and metabolites usually analyzed and measured by nuclear magnetic resonance (NMR) spectroscopy or liquid chromatography tandem mass spectrometry (LC-MS), proteins mainly explored and measured by proteomics and enzyme-linked immunosorbent assay (ELISA), and nucleic acids (RNA and DNA) analyzed by next generation sequencing (NGS) or droplet digital polymerase chain reaction (ddPCR) [9, 15–18]. In the case of the hepatobiliopancreatic (HBP) tumors, due to their anatomical and physiological characteristics, bile may be in direct contact with the lesions, and therefore higher concentrations of tumor biomarkers may be present in bile compared to plasma or urine, increasing the sensitivity of this LB strategy [19]. Moreover, its relatively confined nature and location reduce the possibility of detecting aberrant biomarkers coming from another diseased organ, increasing the specificity of the results. In fact, in this context, as a tumor-adjacent fluid, several studies have shown that the analysis of bile outperforms that of blood, and in addition has the potential to recapitulate tumor heterogeneity [20–23].

However, the use of bile as LB matrix has some limitations, for instance, its availability. LB strategies are generally minimally invasive, however, most methods used to obtain bile (see below) cannot have this consideration. Nevertheless, accumulating evidence demonstrates that patients undergoing interventions for therapeutic or diagnostic purposes in which the biliary tract is accessible can benefit from bile analysis. For instance, and as discussed below, the mutational status of bile cfDNA can help to diagnose strictures of unknown etiology, identifying and anticipating the presence of malignancies [20]. Thus, given the informative potential of bile, its collection should be considered when procedures in which the biliary tract is accessed are performed. Accordingly, the Nouvelle-Aquitaine, Euskadi, and Navarre Euroregion funded a project for the creation of a bile biobank named Bilebank (https://bilebank.org/) where bile and patients’ associated clinical data are available for analytical and research purposes.

Several approaches, surgical, percutaneous (interventional radiology), or endoscopic [24–26] are used in different clinical situations for biliary drainage (Table 1). In those situations, bile could be collected and used to perform different analyses to help with the diagnosis, prognosis, and treatment of patients.

There are multiple pathological conditions (Table 2) in which bile might be collected and its analysis could be clinically relevant (Table 3) [16, 28]. However, in most of these situations, bile composition has not been extensively interrogated, and therefore the utility of measuring changes in bile acids, metabolites, microbiota, proteins, or nucleic acids remains unknown (Figure 1).

However, different studies already support bile analysis as a valuable tool. Indeed, changes in bile composition have been correlated with clinical severity, disease progression, and final outcome in patients with different hepatobiliary diseases including choledochal cysts, extrahepatic portal venous obstruction, and infantile obstructive cholangiopathy [28]. Moreover, changes in bile composition have been proposed to help predict hepatic function and liver failure following liver surgery. For instance, monitoring the levels of interleukin 6 (IL-6) in bile could help detect acute rejection after liver transplantation [71] or liver failure after resection [32], and the sequential monitoring of bile salt composition has been suggested to be useful to discriminate functional versus non-functional grafts during liver transplantation [72].

Biliary crystals are associated with gallstone disease, idiopathic pancreatitis, sphincter of Oddi dysfunction, unexplained biliary pain, and post-cholecystectomy biliary pain [73]. Microscopic bile examination, especially when EUS is not available, can be useful for the diagnosis of these pathologies.

During cholangitis and biliary infections, bile culture is required to guide antimicrobial choice as in intraoperative bile cultures of patients undergoing pancreaticoduodenectomy or total pancreatectomy [38, 45, 74].

Biliary excretion evaluation is another field where bile collection would be of interest, as bile is an important route of elimination for many drugs, with a relevant impact on pharmacokinetics and drug-drug interactions [67, 75].

Further studies are required to elucidate whether bile analysis could also help to diagnose or to identify the mechanisms of disease in patients with jaundice, for instance, alterations of bilirubin metabolism or hepatocellular dysfunction [65]. In the case of obstructive jaundice, bile LB has demonstrated its extraordinary performance in the discrimination between benign and malignant strictures caused by BPT [20]. Although a minority of biliary strictures are benign [76], their accurate diagnosis is a dilemma and the sensitivity of current diagnostic methods does not exceed 60% [77]. As discussed below, the analysis of mutations in the cfDNA isolated from bile obtained during ERCP in patients with biliary strictures can diagnose the presence of BPT with 90% sensitivity [20], without the need for further explorations, and shortening not only the time for diagnosis but for oncological therapy. BPT have a very poor prognosis because they are often diagnosed in advanced stages, mainly due to nonspecific symptoms, and have ineffective oncological treatments. Therefore, biliary molecular analyses would benefit the management and outcome of these patients. In this context, the increased understanding of carcinogenic mechanisms in BPT is leading to new molecular classifications that facilitate the selection of targeted treatments and appropriate prognostic guidance. These molecular studies are commonly conducted on histological samples [78–81]. The use of bile or pancreatic juice obtained during ERCP, or collected after secretin-induced secretion [9, 63] in addition to enhancing diagnostic sensitivity might also guide treatment. Similarly, for biliary strictures developing in patients with potentially preneoplastic conditions (such as PSC, chronic pancreatitis, or pancreatic cysts) requiring ERCP, molecular analyses of bile obtained during the procedure can increase the diagnostic sensitivity for malignancy in cases where the anatomopathological study was inconclusive, or even anticipate the presence of tumors, which may impact transplant decisions [82].

Informative analytes present in bile include bile acids, lipids, metabolites, proteins, cell-free DNA, cell-free microRNAs, extracellular vesicles, and circulating tumor cells. Different recent publications have reviewed the utility of these analytes, extracellular vesicles, and cells present in the bile [5–7, 9]. In the last section of this review, data regarding the identification of mutations and DNA methylation marks in bile cfDNA and the use of bile as a source material for the generation of organoids are summarized.

Many studies and other published evidence, including clinical trials, are demonstrating the potential of LB strategies in the management of patients mainly in the oncology field. Although several biofluids are initially available as LB matrices, most studies are based on blood, due to its ease of collection and informative potential. However, other biofluids such as bile, given its complex and specific composition, which has been shown to vary in pathological conditions, and its anatomical location, in a confined and lesion adjacent space, can provide increased sensitivity and specificity in different clinical situations.

Although much remains to be done, evidence compiled in this review demonstrates the potential of bile analysis to provide molecular information useful for the diagnosis, prognosis, and therapeutic guidance of different diseases, the outcome of pharmacological studies, or the establishment of cellular disease models and regenerative medicine strategies.

In this scenario, bile collection could be implemented in clinical practice in those patients undergoing procedures in which the biliary tract is exposed or accessible. Furthermore, it is not unreasonable to imagine that new strategies might be developed to collect bile in those patients in whom it could be informative to improve their management.

In the near future, standardized LB protocols and sensitive methodologies will revolutionize the management of cancer patients. Bile analysis may be part of this revolution and contribute also to the management of other diseases.