Molecular diagnosis of intrahepatic cholangiocarcinoma

Abstract

Intrahepatic cholangiocarcinomas (iCCA) are primary intrahepatic malignancies originating from biliary epithelia. While both hepatocellular cancer and iCCA can present as mass lesions within the liver, these cancers are distinct in their morphology, etiology, pathology, natural history and response to therapy. There is a need for accurate and sensitive molecular markers for the diagnosis of iCCA. Recent advances in elucidating molecular and genetic characteristics of iCCA offer the potential of molecular-based diagnosis of iCCA. Specific genetic mutations of IDH1/2, BAP1, p53, and KRAS, FGFR gene fusions and alterations in microRNA have all been described in iCCA. Although there are no accurate serum or biliary biomarkers currently available for diagnosis of iCCA, several potential candidates have been identified. Knowledge of specific genetic or molecular abnormalities offers potential for individualized approaches for the treatment of patients with iCCA in the future.

INTRODUCTION

Cholangiocarcinomas are rare malignant tumors arising from the biliary tract. These malignancies are aggressive and are associated with a very poor prognosis. Diagnosis of cholangiocarcinoma requires consideration of the clinical scenario, imaging studies, tumor markers and histologic evaluation. Based on anatomical locations, cholangiocarcinoma can be separated into three distinct tumor types namely intrahepatic cholangiocarcinoma (iCCA), perihilar cholangiocarcinoma (pCCA), and distal cholangiocarcinoma (dCCA). Despite their common association with the biliary tract, these types of cholangiocarcinomas differ in their clinical presentation.

The separation of these cancers into intrahepatic and extrahepatic cancers is best avoided because it has contributed to a lack of clarity in epidemiological, clinical and molecular studies as a result of inconsistent designation of pCCA as either intrahepatic or extrahepatic.

In this review, we will focus on emerging approaches for the molecular diagnosis of iCCA. iCCA are primary cancers of the liver that originate from the intrahepatic biliary ductal system, and form an intrahepatic mass whereas pCCA are tumors located in the bifurcation of the right and left bile duct, and dCCA are tumors originating from the distal part of the extrahepatic bile duct beyond the attachment of the cystic duct. Whereas patients with dCCA or pCCA may present with unremitting jaundice and cholestasis with pruritus, pale stools and dark urine, obstructive symptoms or cholangitis are rare as a symptom of iCCA, except in advanced cases (1). In contrast, iCCA can be identified incidentally as mass lesions on imaging studies that are performed for investigation of other symptoms. In this context, they need to be distinguished from other benign and malignant hepatic lesions.

Populations at risk for developing iCCA

The peak incidence for iCCA is between ages 55 and 75 years. Unlike HCC, which is 5 to 6 times more prevalent in men, iCCA appears to have only a slight male predominance, with a male: female ratio of 2:3(2).

Chronic biliary tract inflammation is an established risk factor for iCCA. Specifically, primary sclerosing cholangitits (PSC), intrahepatic lithiasis, and parasite infections such as Clonorchis sinensis and Opistorchis viverrini are well-established risk factors for iCCA(3). Congenital abnormalities of the biliary system carry a risk of malignant transformation that can be as high as 15% after the second decade of life. Such abnormalities include fibrocystic liver disease, choledochal cysts, and Caroli’s disease. Recently, chronic hepatitis infection has been recognized as a risk factor for iCCA(4). In addition to diabetes mellitus, IBD, and smoking, cirrhosis and hepatitis exposure are also risk factors for iCCA(5). In a recent meta-analysis of published case-control studies, we identified several risk factors for iCCA that included hepatitis B, hepatitis C, obesity, diabetes and alcohol(6). Exposure to certain toxins is also associated with iCCA. Several cases were identified amongst workers at a printing company in Osaka(7). All patients were exposed to 1,2-DCP for 7–17 years and diagnosed with cancer 7–20 years after their first exposure. An increased risk of iCCA has been associated with exposure to thorotrast (thorium dioxide) (3).

Genetic and molecular characteristics of iCCA

Most studies of genetic or molecular features of cholangiocarcnoma have not systematically defined these events in the three clinically defined types of cholangiocarcinomas even though these cancers differ in their presentation, biological behavior and management. Alterations in gene and protein expression that are specific for iCCA are now becoming recognized. Isocitrate dehydrogenase(IDH) mutations were more frequently observed in iCCA than in extrahepatic cancers(8, 9). The overexpression of p53 was not identified in BilIN lesions and was less frequent in iCCA (18.2%) compared with extrahepatic cancers (38.1%) or gall-bladder cancers (61.5%) (10). Chang et al showed that EGFR mutation was an independent prognostic marker in CCA in addition to tumor stage and differentiation. No simultaneous EGFR and KRAS mutations were found in extrahepatic cholangiocarcinoma and gallbladder carcinoma (11). Genetic mutations that have been reported in iCCA are listed in Table 1. Understanding specific genetic and molecular characteristics that are specific for iCCA when compared to other types of cholangiocarcinoma will be essential in order to develop new or more effective diagnostic approaches.

Table 1.

Genetic mutations in intrahepatic cholangiocarcinoma

Mutation	location	Type of alteration	Association	Freguency	reference
E-cadherin	iCCA	Activating mutations	DH	12%	(73)
EGFR	iCCA	Activating mutations	OS	0–15%	(74–76)
GNAS	iCCA	Activating mutations	OS	7.4%	(29)
IDH1/2	iCCA	Inactivating mutations	OS (after surgery)	10–28%	(8, 9, 12)
K-ras	iCCA	Activating mutations	OS	8–54%	(24–27)
BRAF	iCCA	Activating mutations	OS, LNM, ATS	1–22%	(74, 75, 77)
p53	iCCA	Inactivating mutations	OS, LNM, ATS	21–37%	(10, 31, 78)

DH, Differentiation histology; OS, Overall survival; LNM, Lymph node metastasis; ATS, Advanced tumor stage

IDH1/2 mutations

Mutations in the genes encoding isocitrate dehydrogenase, IDH1 and IDH2, have been reported in 10–28% of cholangiocarcinomas(8, 9, 12, 13). They have also been reported in several other types of cancers such as gliomas, myeloid leukemias, chondrosarcomas and thyroid cancer(14). These mutations result in elevated levels of an oncometabolite, 2-hydroxyglutarate, which is associated with higher DNA CpG methylation and altered histone methylation that accompany a block in cellular differentiation(15). Mutations in IDH1 or IDH2 were associated with longer overall survival and were independently associated with a longer time to tumor recurrence after iCCA resection in multivariate analysis. IDH1 and IDH2 mutations were significantly associated with increased levels of p53 in iCCA, but no mutations in the p53 gene were found, suggesting that mutations in IDH1 and IDH2 may result in p53 activation(12).

BAP1 mutations

The BAP1 protein is a deubiquitinase belonging to the ubiquitin C-terminal hydrolase family and is involved in chromatin remodeling. In a complex with ASXL1, BAP1 deubiquitinates histone H2A(16). Somatic mutations in BAP1 occurred in 8 of 32 iCCA (25%)(17). BAP1 was frequently mutated in CCA cases without O. viverrini infection(18). Most of the observed BAP1 mutations were predicted to result in loss of the HCF-1–binding domain and nuclear localization signal, both of which are necessary for the inhibition of cell proliferation and tumorigenesis(16).

FGFR2 fusions

Recent studies have reported the presence of FGFR fusions in cholangiocarcinoma. Wu et al. detected a novel kinase fusion, FGFR2-BICC1, in two cholangiocarcinoma patients(19). Moreover Arai et al. detected FGFR2 fusion in nine of 66 patients (13.6%) with iCCA, of which seven patients had FGFR2-AHCYL1 fusion and two patients had FGFR2-BICC1 (20). Ross et al. reported three gene fusion, FGFR2-BICC1, FGFR2-KIAA 1598 and FGFR2-TACC3, in primary iCCA (21). FGFR gene fusion positive cancers have been shown to have enhanced susceptibility to FGFR inhibitors over activating point mutations of FGFR(19, 20). Borad et al., reported three patients possessing an FGFR2 gene fusions (FGFR2-MGEA5, FGFR2-TACC3 and FGFR2-BICC1), of which two patients received targeted therapy for FGFR2 (22). A patient with a FGFR2-MGEA5 fusion was treated with ponatinib monotherapy with anti-tumor activity documented. In another patient with a FGFR2-TACC3 fusion with progression on pazopanib, stable disease was noted with treatment using a pan-FGFR inhibitor ponatinib (22). Of note, tyrosine kinase inhibitors that target anaplastic lymphoma kinase (ALK) are particularly effective in the treatment of a distinct subset of lung adenocarcinoma carrying ALK fusions(23). These emerging results regarding FGFR2 fusions suggest that oncogenic activation of FGFR2 may represent a therapeutically actionable event and that identification of FGFR2 fusions may have therapeutic implications.

Other genetic changes

Gain-of-function mutations in KRAS downstream of EGFR represent one of the most frequent mutations found in iCCA (8–54%) (24–27). Moreover, mutations in KRAS were detected in 30% of bile from patients with primary sclerosing cholangitis (PSC), suggesting that it is an early event contributing to the malignant transformation of cholangiocytes(28). KRAS were associated with poor overall survival in all patients with iCCA (29).

The p53 tumor suppressor gene is localized on chromosome 17p13 and encodes a 53-kD nuclear phosphoprotein. It plays a central role in DNA repair and apoptosis, thus regulating epithelial cell homoeostasis(30). More than 90 different mutations have been described in TP53. A review of 10 studies which included 229 cases found a total of 21% (49 patients) with mutations in TP53, ranging from 23% in Asia, 14% in Europe, and 26% in the USA(31).

Analysis of mutations was reported in a large set of iCCA patients by Robertson et al. Of 54 cases, KRAS mutations were present in 7.4%, and BRAF mutations in 7.4%; these were mutually exclusive. Mutant cases were associated with a higher tumor stage at time of resection and a greater likelihood of lymph node involvement. These cases were also associated with a worse long-term overall survival(32).

Exome sequencing of eight CCA with Opisthorchis viverrini (Ov) infection and 15 CCA without Ov infection identified a total of 245 somatic mutations in 224 genes. These analyses identified the mutation of TP53, KRAS2, SMAD4, and CDKN2A as frequently mutated genes similar to other prior reports(33).

microRNAs

Alterations in microRNA are associated with modulation of tumor cell proliferation, alteration in sensitivity to chemotherapy, or alteration in epithelial–mesenchymal transition (EMT). Upregulation of microRNA such as miR-26A, miR-31, miR-21, and miR-421 as well as downregulation of microRNAs such as miR-494, miR-370, and miR-138 has been reported in CCA tissues and can modulate cell proliferation through several different mechanisms (34–37). Similary, miRNA associated with chemotherapy resistance include miR-21 which is up-regulated or mir-320 which is down-regulated in CCA. miR-21 can modulate gemcitabine-induces apoptosis by phosphatase and tensin homolog deleted on chromosome 10(PTEN)-dependent activation of PI 3-kinase signaling(34). During EMT, epithelial cells lose their cell polarity and cell–cell adhesion and can gain migratory and invasive properties to become mesenchymal cells(38). The activation of miR-200c can lead to a reduction of EMT with reduced cell migration and invasion in iCCA cells(39). Likewise, miR-214 may be important in regulating metastasis of iCCA because downregulation of this miRNA can promote EMT by directly targeting the Twist gene (40).

Proteins

Many proteins have been associated with iCCA (Table 4). Biological processes that have been implicated in tumorigenesis include EMT, which contributes to invasion and metastasis (41). An EMT phenotype resulting from expression of E-cadherin, β-catenin, vimentin and fibronectin has been reported in iCCA, and MET associated proteins such as CD151, GATA6, Gli1 and S100A4 have been postulated as biomarkers. Korita et al reported a positive correlation of vimentin expression with differentiation and poor survival in iCCA(42). Fibronectin expression occurs in many carcinomas and has been correlated with poor survival and metastasis(43). A study by Gu et al. reported that the combination of β-catenin negativity with positive expression of vimentin or fibronectin showed worse prognosis whereas β-catenin positivity with loss of vimentin or fibronectin expression showed the best prognosis (44).

Table 4.

Candidate Markers in tissues

gene/protein	location	poor prognosis	compartment	method	Association	reference
Fibronectin	iCCA	positive	membranous	IHC	LNM, ATS	(44)
vimentin	iCCA	positive	membranous	IHC	ATS, DH	(44, 83, 84)
E-cadherin	iCCA	negative	membranous	IHC	LNM, ATS, DH, OS, NI	(44, 83–86)
β catenin	iCCA	negative	membranous	IHC	TS, DH, LNM	(44)
Fibronectin-β catenin	iCCA	positive/negative	membranous	IHC	LNM, OS	(44)
vimentin-β catenin	iCCA	positive/negative	membranous	IHC	DH, NI, LNM, ATS, OS	(44)
vimentin-E cadherin	iCCA	positive/negative	membranous	IHC	DH, NI, RMI, ATS	(44)
GATA6	CCA	positive	Nuclear	IHC	OS, OS (after surgery), LNM	(46)
Gli 1	iCCA	positive	Nuclear	IHC	OS (after surgical)	(47)
Capn4	iCCA	positive	cytoplasmic	IHC(Tissue microarray)	OS (after surgery), LNM, ATS	(87)
DKK1	iCCA	positive	cytoplasmic	IHC	OS (after surgery), ATS	(61)
Fascin	iCCA	positive	cytoplasmic, membranous	IHC	DH, VI, LNM, DM, OS	(83, 88)
IL-17	iCCA	positive	cytoplasmic	IHC(Tissue microarray)	OS (after surgery)	(89)
MUC16	iCCA	positive	cytoplasmic	IHC	OS (after surgery)	(90)
N-cadherin	iCCA	positive	membranous	IHC	VI	(84)
p-4EBP1	iCCA	positive	cytoplasmic	IHC	OS	(91)
p-AKT1	iCCA	negative	cytoplasmic	IHC	OS	(92)
PDGF	CCA	negative	cytoplasmic	IHC	OS, ATS, DM	(93)
p-mTOR	iCCA	negative	cytoplasmic, membranous	IHC	OS	(91)
PTEN	iCCA	negative	cytoplasmic	IHC	OS	(92)
Smad7	CCA	positive	cytoplasmic	IHC	LNM, NI, OS, OS (after surgery)	(85)
CD151	iCCA	positive	membranous	IHC(Tissue microarray)	OS, OS (after surgery), DH, VI, LNM	(94)
S100A4	CCA	positive	Nuclear	IHC	OS, OS (after surgery)	(49)
Beclin1	iCCA	negative	cytoplasmic	IHC	LNM, OS, OS (after surgery)	(95)
MAGE-A3/4	iCCA	positive	cytoplasmic	IHC	OS	(96)
c-Met	iCCA	positive	cytoplasmic	IHC	LNM, DH, OS, OS (after surgery)	(94)
EGFR	CCA	positive	membranous	immunohistochemistry FISH	OS, OS (after surgery)	(97, 98)
Periostin	iCCA	positive	stromal compartment	IHC	OS	(99)
PRL-3	iCCA	positive	cytoplasmic	IHC	ATS, VI, LNM, OS	(100)
snail	CCA	positive	membranous	IHC	OS (after surgery), LNM	(101)
IMP3	CCA	positive	cytoplasmic	IHC(Tissue microarray)	OS	(102)
p27	iCCA	negative	Nuclear	IHC	OS (after surgery), LNM	(103)
Skp2	iCCA	positive	Nuclear	IHC	OS (after surgery)	(103)
CD44	CCA	positive	membranous	IHC	OS, ATS	(104)
P120-catenin	iCCA	negative	membranous	IHC	DH, ATS, OS	(86)
p16	CCA	negative	Nuclear	IHC(Tissue microarray)	OS	(105)
VEGF-C	iCCA	positive	cytoplasmic	IHC	OS (after surgical), LNM	(106)

LNM, Lymph node metastasis; DH, Differentiation histology; OS, Overall survival; ATS, Advanced tumor stage; VI, Vascular invasion; NI, Neural invasion; TS, Tumor size; DM, Distant metastasis; IM, Intrahepatic metastases; RMI, Resection margin involve; IHC, Immunohistochemistry; FISH, fluorescence in situ hybridization

Recent studies have demonstrated that CD151 forms structural and functional associations with the proto-oncogene that encodes c-Met protein and is involved in the regulation of downstream pathways of the c-Met/hepatocyte growth factor (HGF) system(45). Huang et al. examined the expression of CD151/c-Met by immunohistochemistry in a tissue microarray. Overexpression of CD151 was implicated in metastasis and invasion of iCCA. Thus, either CD151 or c-Met overexpression may be potential molecular therapeutic targets for iCCA. Tian et al. reported that 67LR was regulated by GATA6 through binding to its promoter in CCA cells, indicating that aberrant expression of GATA6 correlates with poor prognosis and promotes tumor cell invasion and metastasis, possibly through promoter binding mediated regulation of 67LR (46). In iCCA tissues, Gli1 nuclear immune-intensity is associated with intrahepatic metastasis and venous invasion(47). Gli1 is a transcriptional target of the hedgehog pathway, and Gli1 expression serves as an indicator of activated hedgehog signaling. Elevated Gli1 expression was linked with cancer development and progression(48). Furthermore, blocking hedgehog signaling by cyclopamine or siRNA-targeting Gli1 results in apoptosis and growth inhibition in iCCA cells. Thus, detection of hedgehog pathway activation may have therapeutic value(47). S100A4 may also represent a potential therapeutic target. Fabris et al. reported that nuclear expression of S100A4 by neoplastic ducts in CCA was a strong predictor of metastasis and reduced survival after resection. S100A4-silenced EGI-1 cells (human CCA cell line) demonstrated reduced motility, invasiveness, and MMP-9 secretion in vitro, without changes in cell proliferation(49).

Molecular diagnosis of iCCA

Serum biomarkers

At this time there are no available serum biomarkers that are useful for early or accurate diagnosis of iCCA. Both CA 19-9 and carcinoembryonic antigen (CEA) can be elevated in these cancers but are not specific and may also be elevated in other cancers or in the setting of cholestasis in the absence of malignancy, and following liver injury (50). Thus, their accuracy for the diagnosis of cholangiocarcinoma is limited. Other potentially useful markers include DU-PAN-2, CA 125 and interleukin-6(51). CYFRA 21-1 concentrations again were shown to be significantly higher in CCA than in HCC or in patients with benign liver diseases. Moreover, sensitivity to detect CCA was superior for CYFRA 21-1 as compared with alpha fetoprotein, CEA, and CA 19-9 levels (87% versus 17, 35, and 61%, respectively)(52). Combinations of these markers might improve the sensitivity and specificity for the diagnosis of cholangiocarcinoma.

Several other several candidate serum marker for iCCA have been reported, including KL-6 mucin(53), hTERT mRNA(54), A1BG/AFM ratio(55) and MUC5AC(56). Candidate markers in serum are listed in Table 2. Matsuda et al. reported that Wisteria floribunda agglutinin (WFA) can differentiate iCCA lesions from normal bile duct epithelia (57). Using serum as well as bile from patients with CCA or benign bile duct diseases, they reported the use of WFA-positive L1CAM, enriched from serum by the WFA-assisted affinity capturing, to distinguish malignant from benign epithelia. Diagnostic accuracy for CCA was increased by combining the assay with a high sensitive assay detecting WFA-positive sialylated mucin 1 (overall accuracy = 0.84, AUC = 0.93)(58). Shen et al. reported that SSP411 (also known as spermatogenesis-associated protein 20), was a potential serum diagnostic biomarker for CCA, with a sensitivity of 90.0% and specificity of 83.3% at a cutoff value of 0.63(59). SSP411, a thioredoxin family member, is a novel spermatid-expressed gene which is thought to play a role in sperm maturation, fertilization and/or embryo development(60). In another report, serum DKK1 levels were significantly higher in iCCA patients than in healthy volunteers. On ROC analyses, a serum DKK1 level of 2.49 ng/mL had an area under the curve = 0.872, with 75.7% sensitivity and 100% specificity for diagnosis of iCCA. High DKK1 expression in iCCA tissues was associated with elevated matrix metalloproteinase 9 (MMP9), vascular endothelial growth factor C (VEGF-C) expression, and high lymph node metastasis(61).

Table 2.

Candidate Markers in serum

Gene/protein	location	material	method	Association	reference
DKK1	iCCA	serum	ELISA	OS	(61)
WFA-L1CAM	iCCA	serum, bile	ELISA	Diagnosis	(57, 58)
A1BG/AFM ratio	CCA	serum	Western blot	Diagnosis, OS (after surgery)	(55)
SSP411	CCA	serum, bile	ELISA, Western blot	Diagnosis	(59)
MMP7	CCA	serum	ELISA	Diagnosis	(79)
CYFRA21-1	iCCA	serum	ECLIA	LNM, ATS, VI, IM	(80)
KL-6 mucin	iCCA	serum	ELISA	Diagnosis,	(53)
MUC5AC mucin	CCA	serum	sandwich ELISA	Diagnosis	(56)
hTERT mRNA	CCA	serum	Real time RT-PCR	Diagnosis	(54)
RCAS1	CCA	serum	ELISA	Diagnosis	(81)

OS, Overall survival; LNM, Lymph node metastasis; ATS, Advanced tumor stage; VI, Vascular invasion; IM, Intrahepatic metastases; ELISA, Enzyme-Linked ImmunoSorbent Assay

Bile biomarkers

Candidate markers in bile are reported in Table 3. The frequency of CA 19-9 detection in bile from patients with both benign and neoplastic diseases of the pancreaticobiliary tract ranges from 46 to 61% and specificity of 60 to 70%(51). However, a higher frequency of elevated CA 19-9 in bile has been reported in patients with acute cholangitis as compared with CCA(62). Similarly, the results regarding the diagnostic value of CEA levels in bile are contradictory. The sampling time as measured before and after biliary drainage procedures has been shown to influence the CA 19-9 and CEA levels, and is a major variable that influences the value of bile biomarkers.

Table 3.

Candidate Markers in bile

Gene/protein	location	material	method	Association	reference
WFA-L1CAM	iCCA	serum, bile	ELISA	Diagnosis	(57, 58)
SSP411	CCA	serum, bile	ELISA, Western blot	Diagnosis	(59)
Mac-2BP	CCA	bile	ELISA	Diagnosis	(82)

ELISA, Enzyme-Linked ImmunoSorbent Assay

S100 protein has been reported as a candidate bile biomaker(63). The bile levels of S100P were increased significantly in patients with cholangiocarcinoma compared with those in patients with lithiasis(63). Moreover, Baraniskin et al. reported that measurement of RNU2-1f levels in bile fluids enabled the differentiation of patients with CCA from controls in all cases. Furthermore, RNU2-1f levels in bile fluids of patients with CCA were significantly higher than in patients with PSC(64).

Li et al. (65) reported that analysis of microRNA within extracellular vesicles obtained from bile enabled diagnosis of CCA. A panel of 5 microRNAs (miR-191, miR-486-3p, miR-1274b, miR-16 and miR-484) demonstrated a sensitivity of 67% and specificity of 96%. Importantly, their control group contained 13 PSC patients, 16 patients with biliary obstruction of varying etiologies (including benign biliary stricture, papillary stenosis, choledocholithiasis, extrinsic compression from pancreatic cysts, and cholangitis), and 3 patients with bile leak syndromes. Their findings establish the importance of using extracellular vesicles, rather than whole bile, for developing disease biomarkers (66).

Tissue diagnosis

Distinguishing hepatic metastases from distant adenocarcinoma and intrahepatic cholangiocarcinoma is difficult because of the similarities in morphological appearances. Current approaches to diagnosis require the exclusion of other primary malignancies. Tissue markers with high specificity for iCCA may be helpful for the diagnosis in these difficult cases. In addition, tissue markers that correlate with disease progression, or response to therapy would be of particular clinical value in defining prognosis or future care. Selected candidate tissue markers along with their clinical or pathological correlations are shown in Table 4. Finally, specific therapies may also be suggested by demonstration of drug-relevant targets within tissues.

Therapeutic options for iCCA

For patients with localized disease, surgical resection should be considered as this is the only strategy with the potential for cure. A critical factor is the extent of hepatic resection that is necessary and also compatible with a functionally adequate remnant liver. Advances in techniques to predict residual tumor volume and function, and portal embolization enable more extensive resections than were possible previously(67). For resections with either positive margins or residual tumor or positive lymph nodes, re-resection or ablation should be considered if feasible(68). Locoregional approaches, such as ablation, transarterial chemoembolization (TACE) and transarterial radioembolization have been used in patients with cholangiocarcinoma(69). However, the experience has been limited and these modalities have not been systematically evaluated. Although the response to external beam radiation therapy in patients with intrahepatic cholangiocarcinoma has been poor, anecdotal evidence indicates that a reduction in tumor burden can be achieved with stereotactic body radiotherapy treatment (SBRT)(70). For patients with advanced cancers that are unresectable or those with metastatic disease, systemic therapy with gemcitabine plus cisplatin is a first-line approach(71). Alternative choices, particularly in individuals who might not tolerate the combination of gemcitabine and cisplatin would include gemcitabine monotherapy, 5FU-based regimens, or supportive care (72). Future directions in therapy of iCCA are likely to be based on the emerging knowledge of the molecular pathogenesis of these cancers. Drugs that can target many of the oncogenic pathways that have been identified such as FGFR2 fusions, IDH mutations, EGFR and others are available, and may form the basis for targeted individualized therapy in the future.

Conclusions

There is a need for accurate and sensitive molecular markers for iCCA. Similarly, a knowledge of specific genetic or molecular abnormalities offers the potential for more refined and individualized approaches to the treatment of patients with iCCA through the use of appropriate targeted therapies. Understanding genetic and molecular changes associated with these cancers therefore holds the promise for future assays that may be helpful for disease diagnosis, prognosis, or treatment.