Abstract
Cholangiocarcinoma (CCA) is the second most common primary hepatic malignancy worldwide. The incidence of intrahepatic CCA is increasing, whereas that of extrahepatic CCA is decreasing. This review looks at the new advances that have been made in the management of CCA, based on a PubMed and Science Citation Index search of results from randomized controlled trials, reviews, and cohort, prospective and retrospective studies. Aggressive interventional approaches and new histopathological techniques have been developed to make a histological diagnosis in patients with high risk factors or suspected CCA. Resectability of the tumour can now be assessed using multiple radiological imaging studies; the main prognostic factor after surgery is a histologically negative resection margin. Biliary drainage and/or portal vein embolization may be performed before extended radical resection, or liver transplantation may be undertaken in combination with neoadjuvant chemotherapy or chemoradiotherapy. Though many advances have been made in the management of CCA, the standard modality of treatment has not yet been established. This review focuses on the clinical options for different stages of CCA.
Introduction
Cholangiocarcinoma (CCA) is a fatal cancer of the biliary epithelium; it arises either within the liver (intrahepatic cholangiocarcinoma; ICC) or in the extrahepatic bile ducts (extrahepatic cholangiocarcinoma; ECC). Globally, CCA is the second most common primary hepatic malignancy, with a reported incidence of one to two cases per 100 000 in the USA. 1 Several epidemiological studies have shown that the incidence and mortality rates of ICC are increasing, while those of ECC are falling.2–7
The exact aetiology of CCA is unknown. There are several well-defined risk factors, however, including primary sclerosing cholangitis, liver fluke infestation, congenital fibropolycystic liver disease and intrahepatic biliary stones.2,5,8,9 Other risk factors include exposure to dioxin, Thorotrast or nitrosamines. 10
As there are no specific symptoms in early malignant lesions, patients with CCA mostly present in the advanced stages of the disease, which contributes to its poor prognosis. With the advent of new techniques such as intraductal ultrasonography and
This review was based on literature searches in PubMed and the Science Citation Index using the following search terms: cholangiocarcinoma and diagnosis; aetiology; surgery; extended surgery; liver transplantation; neoadjuvant chemoradiation; chemotherapy; radiotherapy; palliative biliary drainage; and photodynamic therapy. Results from reviews, case reports, randomized controlled trials, and cohort, prospective and retrospective studies for which the title and abstract were available in English were included. Studies with <10 patients were excluded.
Diagnosis
The clinical features of CCA depend on the stage and location of the tumour. As there are no specific symptoms in the early stages of CCA, most patients present at an advanced stage. Patients with ECC usually present with obstructive jaundice, whereas those with ICC usually present with abdominal pain. Common complaints include pruritus, weight loss, fever, and symptoms related to biliary obstruction such as clay-coloured stools and dark urine. Physical signs include jaundice, hepatomegaly and a right upper quadrant mass. Serum alkaline phosphatase and bilirubin levels are elevated if bile duct obstruction is present. No specific tumour markers have yet been identified in CCA. Although the sensitivity and specificity of carcinoembryonic antigen and cancer antigen 19-9 (CA19-9) are low, they may be of value in predicting prognosis after surgery or for screening of patients, especially in those with predisposing risk factors.11,12 Other markers may have clinical significance. Serum levels of matrix metalloproteinase-713 and tumour M2-PK
14
have been shown to have the potential to differentiate CCA from benign biliary tract disease with more sensitivity and specificity than CA19-9. In addition, an animal trial reported that a novel CCA-associated carbohydrate antigen may have potential as a marker for the early diagnosis of CCA.
15
Zabron
In patients with suspected CCA, transabdominal ultrasonography and other non-invasive imaging should be performed to confirm the diagnosis. Transabdominal ultrasound is sensitive for visualizing the bile ducts, confirming ductal dilatation and ruling out choledocholithiasis. For precise characterization of the neoplasm and planning further management, however, other imaging modalities such as computed tomography (CT), contrast-enhanced CT (including three-dimensional reconstruction, three-phase CT and CT angiography), cholangiography, positron emission tomography (PET) and magnetic resonance imaging (MRI), including magnetic resonance cholangiopancreatography (MRCP), should be carried out preoperatively.
Computed tomography and contrast-enhanced CT can not only visualize the local anatomical structures, measure the size of the tumour and the extent of the bile duct dilatation, and detect regional lymph node enlargement, atrophy of the lobe and satellite nodules, but also have the advantage of being able to perform precise multidirectional assessment of biliary and vascular involvement, which helps in the accurate prediction of resectability.20–25 Similarly, MRCP in combination with MRI is a reliable non-invasive diagnostic method for the pre-therapeutic staging of CCA. Due to its intrinsic high tissue contrast and multiplanar ability, MRI with MRCP is capable of examining all the structures involved, such as bile ducts, vessels and hepatic parenchyma, and a precise preoperative assessment of the tumour can therefore be achieved.26–29 There are no major differences between CT and MRI for preoperative appraisal for patients with CCA,
30
and these imaging techniques play complementary roles in the process of clinical diagnosis and preoperative assessment. PET using the radiotracer [18F]fluorodeoxyglucose has become a useful staging technique for many neoplasms. One study of 123 patients with suspected and potentially operable CCA demonstrated that PET-CT was more accurate than CT in the diagnosis of regional lymph node metastases (75.9% versus 60.9%,
Although the rapid development of imaging technology and instrumentation has enabled the accurate demonstration of lesions, these imaging modalities are of limited value in early CCA, when there are small or even no changes in morphology. In addition, differentiating between benign and malignant bile duct stricture is very difficult, but this distinction is important in treatment planning. These clinical problems can be addressed by the use of cytology or tissue biopsy via endoscopic retrograde cholangiopancreatography (ERCP), percutaneous transhepatic cholangiography (PTC), cholangioscopy or endoscopic ultrasonography (EUS)-guided fine needle aspiration.
Due to its relative ease and safety, many studies have suggested that cytology during ERCP, despite its low sensitivity, remains a good choice for the diagnosis of causes of biliary stricture.32–35 To improve the sensitivity, further refinements in technique and procedure have been suggested. One study of cytodiagnosis through ERCP showed that intraductal aspiration had a significantly higher sensitivity (89% versus 78% for adequate samples and 89% versus 37% for all samples) and significantly superior cellular adequacy (92.8% versus 35.7%) than brushing in patients with suspected malignant biliary stricture. 36 In patients with negative results on ERCP-guided bile duct biopsy, biopsy sensitivity was improved by the use of intraductal ultrasonography (IDUS)-guided forceps during ERCP. 37 Similarly, cytology or biopsy during PTC has been shown to be effective and safe. In the study of Jung et al., 38 patients with obstructive jaundice underwent transluminal forceps biopsy during or after percutaneous transhepatic biliary drainage (PTBD), with a sensitivity, specificity and accuracy of 78.4%, 100% and 79.2%, respectively, without any major complications related to the biopsy procedures. Other studies have also demonstrated that biopsy during PTBD is a safe procedure and can provide relatively high accuracy in the diagnosis of malignant biliary obstructions.39,40 Currently, there are no significant differences in sensitivity and complications between cytology/biopsy with ERCP or PTC, but more attention should be paid to the fact that catheter tract implantation metastasis is not a rare complication following PTBD in ECC.41,42 The study of Kim et al. 43 indicated that PTC in combination with IDUS was highly accurate for assessing Bismuth type in patients with hilar CCA, which may help in the identification of an optimal surgical plan for the treatment of hilar CCA, especially in Bismuth type III and IV. IDUS images also have important clinical significance in the differentiation of malignant and benign lesions. Tamada et al. 37 reported that when IDUS images showed a polypoid lesion, localized wall thickening, intraductal sessile tumour or sessile tumour outside of the bile duct, the sensitivities of the biopsy were 80%, 50%, 92% and 53%, respectively, and that the presence of sessile tumour (intraductal or outside of the bile duct), tumour size >10.0 mm and interrupted wall structure on IDUS images could predict malignancy in patients with a negative ERCP-guided bile duct biopsy. In addition, a number of studies have demonstrated the safety and high accuracy, sensitivity and specificity of EUS and EUS-guided fine needle aspiration in patients with negative results after endoscopic brush cytology and biopsy.43–48 These results suggest that these techniques can play a significant role in planning further management.
As patients with primary sclerosing cholangitis have a high risk of developing CCA, attention should be paid to early detection of malignant lesions in these patients. Tumour serology combined with IDUS and cross-sectional liver imaging and cytology during ERCP/PTC have been shown to be helpful for CCA screening and diagnosis in patients with primary sclerosing cholangitis.49–51 Naitoh et al.
52
reported that IDUS findings were useful for distinguishing immunoglobulin G4-related sclerosing cholangitis from CCA. In addition, Huddleston et al.
53
described the use of UroVysion™ fluorescence
Treatment
Resection
Surgical resection is the only potentially curative approach currently available, although distant metastasis to the lung, peritoneum or other organs is a contraindication for resection. Preoperative evaluation of the future remnant liver volume and the patient’s general condition is important in deciding whether or not they are suitable for surgery. The prognosis of patients with CCA after surgery is generally poor, with a reported 5-year survival rate in all patients of <20%, improving to approximately 30% in those with R0 resection; median survival times were 15 and 28 months, respectively. 54 Many studies have demonstrated that the major independent prognostic factor after surgery is R0 resection of the tumour.54–57 To achieve R0 resection, hepatectomy and/or pancreaticoduodenectomy is frequently required. Local resection in combination with caudate lobectomy for hilar CCA has resulted in a greater number of patients with R0 resection and has improved the long-term prognosis. 58 More aggressive resection with trisegmentectomy or even semihepatectomy has been used in some patients and was associated with significantly increased survival.59–60 However, the postoperative mortality due to liver dysfunction is also slightly higher with these more extensive operations.58–61 To reduce the risk of postoperative liver dysfunction with semihepatectomy or resection of more than 50–60% of the liver, some researchers have employed preoperative ipsilateral portal vein embolization, which can induce compensatory hypertrophy of the future remnant liver.62–64 Some patients deemed not suitable for surgery because of liver dysfunction or severe cholangitis secondary to cholestasis or bile duct obstruction may be able to undergo surgical resection following preoperative biliary drainage. 65
Other prognostic factors after R0 resection for CCA include lymph node metastasis, perineural invasion and combined vascular resection due to portal vein and/or hepatic artery invasion.1,54,66–68 In hilar CCA, the invasion depth of the tumour (≥5 mm versus <5 mm) has been reported to be a better predictor of long-term outcome than the American Joint Committee on Cancer staging system. 69 In patients with ICC, the macroscopic histopathology has been shown to be useful for predicting survival after hepatectomy, with the mass-forming plus periductal infiltrating type having a more unfavourable prognosis than the mass-forming type. 59
Liver transplantation
The use of liver transplantation is controversial as CCA has a poor prognosis with high recurrence rates. However, a study at the Mayo Clinic in the USA found that after liver transplantation with neoadjuvant chemoradiation, the survival rate of patients with localized, node-negative hilar CCA was significantly higher than that of patients with resectable hilar CCA after R0 resection (
Palliative biliary drainage
As patients with CCA mostly present at an advanced stage, many are unsuitable for curative resection because of chronic duct obstruction resulting in recurrent cholangitis and liver dysfunction. For these patients, the main purpose of therapy is to relieve symptoms (pain, pruritus, jaundice and cholangitis) and improve their quality of life.
The high success rate and low risk of endoscopic biliary drainage achieved by developments in endoscopic apparatus and technology has encouraged its use in patients with definite unresectable CCA.
72
Compared with endoscopic drainage, percutaneous biliary drainage has a similar success rate and procedure-related risk, but can better image the proximal extent of the tumour.
73
It is often performed when endoscopic drainage has failed, when there is infection of isolated obstructive segments or in the absence of high level endoscopic expertise for complex procedures. The two modalities should be used complementally in the management of biliary obstruction, especially after failure of ERCP.
74
A number of studies have demonstrated that EUS-guided biliary drainage is an effective technique in obstructive jaundice.75–77 As several randomized controlled trials have suggested the superiority of metal stents over plastic stents for patency,78–80 the use of a metal stent is preferred in patients expected to survive for more than 6 months. Besides the choice of the stent, whether one or both lobes of the liver should be drained is also controversial. Drainage of 30% of the liver volume has been shown to be efficacious in the relief of symptoms.
81
A randomized trial demonstrated that unilateral drainage had a higher success rate (88.6% versus 76.9%,
The classic palliative surgery for patients with malignant biliary obstruction is a biliary–enteric bypass. However, surgical drainage procedures show no superiority in terms of procedure-related mortality, survival or cost-efficiency rates over non-surgical drainage procedures such as PTBD, endoscopic nasobiliary drainage and endoscopic or percutaneous stent implantation. 84 Therefore, non-operative biliary drainage should be the first choice to resolve biliary obstruction in non-resectable CCA, with surgical bypass being reserved for patients with failed endoscopic/percutaneous drainage and a good life expectancy. 85
Chemotherapy
Chemotherapy has been used in an attempt to control disease and to improve survival and quality of life in patients with unresectable, recurrent or metastatic CCA. In a retrospective study of 93 patients with unresectable or metastatic CCA, those treated with chemotherapy had a significantly higher median overall survival than those who did not receive chemotherapy (
To reduce the drug toxicity of systemic chemotherapy and improve effectiveness in unresectable CCA, chemotherapeutic agents have been given via transcatheter arterial infusion. van Riel et al. 95 observed that gemcitabine given via a 24-h hepatic arterial infusion was well tolerated and resulted in significantly lower systemic gemcitabine plasma concentrations than intravenous infusion. However, Inaba et al. 96 reported that the toxicity of 1000 mg/m2 gemcitabine via transcatheter arterial infusion in patients with unresectable ICC was tolerable, but the desired efficacy could not be reached. A retrospective study by Kuhlmann et al. 97 reported that progression-free survival and overall survival in transarterial chemoembolization with irinotecan-eluting beads were similar to those for systemic chemotherapy with oxaliplatin and gemcitabine (3.9 months versus 6.3 months, and 11.7 months versus 11.0 months, respectively), but were superior to rates in transarterial chemoembolization with mitomycin-C (progression-free survival of 1.8 months, overall survival of 5.7 months). A further study also found that treatment with transarterial chemoembolization with gemcitabine and cisplatin resulted in significantly longer survival than transarterial chemoembolization with gemcitabine alone (13.8 months versus 6.3 months). 98
Micro-RNA (miRNA) can modulate gene expression. Alterations in miRNA expression lead to tumour response to chemotherapy. The inhibition of miR-21 and miR-200b have been shown to increase sensitivity to gemcitabine in CCA. 6 Suppression of galectin-3 expression in CCA cells with siGal-3-K402 significantly enhanced apoptosis induced by cisplatin or 5-fluorouracil, whereas overexpression of Gal-3 led to an increased resistance to drugs. 99
Radiotherapy
Radiotherapy, including external beam radiotherapy and intraductal radiotherapy, uses high-energy X-rays to damage DNA, resulting in tumour tissue necrosis. In patients with advanced CCA that is unsuitable for curative resection, radiotherapy alone or in combination with other approaches such as chemotherapy and/or biliary decompression is an effective treatment option, prolonging survival and improving quality of life.100–104 Patients with resectable ECC who had microscopically positive resection margins showed higher median disease-free survival rates (21 months versus 10 months,
Chemoradiation therapy
A combination of chemotherapy with radiation should theoretically be more effective than either method alone. Leong et al. 109 demonstrated the additional benefit of chemoradiation in patients with either unresectable or locally advanced CCA. A retrospective study by Nelson et al. 110 also suggested that postoperative chemoradiotherapy had a possible benefit in terms of local control in patients with advanced, resected ECC. Preoperative adjuvant chemoradiotherapy is mainly used in patients preparing for liver transplantation. Selected patients with localized, node-negative, unresectable hilar CCA treated by liver transplantation in combination with neoadjuvant chemoradiotherapy achieved significantly higher 5-year survival rates, lower incidences of tumour recurrence and later recurrence than those of patients with resectable hilar CCA after R0 resection, 70 and orthotopic liver transplant combined with neoadjuvant and adjuvant therapies was associated with better survival rates than transplant alone or transplant with adjuvant therapy in patients with intrahepatic or hilar CCA. 71 Some small studies have also reported an added benefit in patients treated with preoperative chemoradiotherapy before resection.110,111
Photodynamic therapy
Photodynamic therapy is the use of a laser to activate
Conclusion
In conclusion, though many advances have been made in the management of CCA, the standard modality of treatment has not yet been established. This review focuses on the clinical options for different stages of CCA.
Footnotes
Declaration of conflicting interest
The authors declare that there are no conflicts of interest.
Funding
This work was supported by The National Natural Science Foundation of China (grant no. 81171444).
