Cholangiocarcinoma in Italy: A national survey on clinical characteristics, diagnostic modalities and treatment. Results from the “Cholangiocarcinoma” committee of the Italian Association for the Study of Liver disease
Article Outline
- Abstract
- 1. Introduction
- 2. Material and methods
- 3. Results
- 4. Discussion
- Conflict of interest statement
- Acknowledgements
- Appendix A. Survey participants with location of the hospital unit in parenthesis
- References
- Copyright
Abstract
Background
Very few studies assessed cholangiocarcinoma clinical characteristics.
Aim
To evaluate the clinical characteristics of intra-hepatic (IH) and extra-hepatic (EH)-CCA.
Methods
We performed a national survey based on a questionnaire.
Results
218 cholangiocarcinomas were observed (47% EH-CCA, 53% IH-CCA) with an age at the diagnosis higher for EH-CCA. Coexistence of cirrhosis or viral cirrhosis was more frequent in IH-CCA than EH-CCA. An incidental asymptomatic presentation occurred in 28% of IH-CCA vs 4% EH-CCA whilst, 74% EH-CCA vs 28% IH-CCA presented with jaundice. 91% of IH-CCA presented as a single intra-hepatic mass, whilst 50% of EH-CCA was peri-hilar. In the diagnostic work-up, 70% of all cholangiocarcinoma cases received at least 3 different imaging procedures. Tissue-proven diagnosis was obtained in 80% cholangiocarcinoma. Open surgery with curative intent was performed in 45% of IH-CCA and 29% EH-CCA. 18% IH-CCA vs 4% EH-CCA did not received treatment.
Conclusions
In Italy IH-CCA is managed as frequently as EH-CCA. In comparison to EH-CCA, IH-CCA occurs at younger age and is more frequently associated with cirrhosis and with an incidental asymptomatic presentation. In contrast, most EH-CCAs are jaundiced at the diagnosis. Cholangiocarcinoma diagnostic management is cost- and time-consuming with curative surgical treatment applicable more frequently in IH-CCA.
Keywords: Cholangiocarcinoma, Clinical presentation, Diagnosis, Extra-hepatic cholangiocarcinoma, Intra-hepatic cholangiocarcinoma, Risk factors, Treatment
1. Introduction
Cholangiocarcinoma (CCA) is a malignant tumour arising from the neoplastic transformation of cholangiocytes, the epithelial cells lining the biliary tree [1], [2], [3], [4], [5], [6], [7]. From the anatomical point of view, cholangiocarcinoma may by classified into intra-hepatic cholangiocarcinoma (IH-CCA) and extra-hepatic cholangiocarcinoma (EH-CCA), the latter including the peri-hilar form, also called Klatskin tumour [1], [2], [3], [4], [5], [6], [7]. CCA represents worldwide the second most frequent primitive liver malignancy [8], [9], [10], [11], [12], [13], [14]. In addition, over the last two decades, a progressive increase of incidence and mortality for CCA has been reported worldwide and this mainly involves the IH-CCA, whereas the EH-CCA levels seem to be stable or slightly decreasing [9], [10], [11], [12], [15].
CCA is characterised by a very bad prognosis (6–12 months survival) and virtually null response to chemotherapy [1], [2], [3], [4], [5]. The only effective treatment is radical surgical resection which, however, is applicable in only approximately 40% of patients since CCA is mostly diagnosed at an advanced stage [1], [2], [3], [4], [5]. This mainly occurs because, in the majority of cases, CCA is clinically silent, with symptoms only developing at an advanced stage. An additional reason is the lack of effective serum biomarkers to be used for screening purposes [16], [17]. Apart from investigations in single centres [18], [19], [20], very few studies have assessed the demographic and clinical characteristics of CCA patients, including clinical presentation, modalities of diagnosis and treatment strategies. In this study, we present the results of a survey promoted by the “Cholangiocarcinoma” committee of the Italian Association for the Study of Liver disease (AISF) aimed at evaluating the demographic and clinical characteristics of CCA patients.
2. Material and methods
The survey, promoted by the “Cholangiocarcinoma” AISF committee, started in August 2008 and ended in July 2009. During this 12-month period, members of AISF (Italian Association for the Study of Liver diseases) and SIGE (Italian Society of Gastroenterology) were asked by e-mail to fill out a form describing CCA patients recruited in their hospital centres, either as out- or in-patients. The form was structured as a questionnaire with multiple choice questions and contained information about the demographic characteristics, clinical presentation, diagnostic modalities and treatment of CCA. As far as smoking habits and alcohol consumption were concerned, only answer no or yes, past or current use was due. For associated pathologies, participants were asked to indicate one or more of the following: primitive sclerosing cholangitis (PSC), gallstones, past cholecystectomy, past endoscopic sphincterotomy, open surgery of the biliary tree, liver cirrhosis, HCV-Ab positivity, HBsAg positivity and others (which had to be specified). Participant physicians were asked to indicate in months the approximate onset of CCA, based on the appearance of clinical symptoms. As far as clinical presentation was concerned, the presence of one or more of the following had to be indicated: jaundice, fever, weight loss, abdominal pain, others (which had to be specified). The diagnostic modalities were investigated by asking participants to indicate one or more of the following imaging procedures: abdominal ultrasonography (US), magnetic resonance (MR), MR-cholangiopancreatography (MRCP), computed tomography (CT), positron emission tomography (PET), endoscopic US (EUS), endoscopic retrograde cholangiopancreatography (ERCP) and others (which had to be specified). Procedures allowing a definitive pathological diagnosis had to be indicated with an X for the following: percutaneous biopsy, endoscopic biopsy, bile cytology, brushing, laparotomy, ERCP and others with required specifications. Participant physicians were asked to classify CCA as localised or with distant metastasis. For the anatomic classification of CCA, the form included a picture showing the 8 liver segments and the biliary tree, where participants were asked to indicate with an X the exact tumour location. Gallbladder cancer was excluded from the analyses. According to the current classification [1], [2], [3], [4], [5], CCAs were classified as IH-CCA or EH-CCA, the latter further subclassified as peri-hilar (Klatskin), CCA involving the middle common bile duct or the distal common bile duct. The peri-hilar tumours were classified according to the Bismuth classification. Treatment modalities were investigated by asking participants to select one or more of the following options: surgery (which had to be specified), chemotherapics (which had to be specified), bile drainage (which had to be specified), radiotherapy (which had to be specified), photodynamic therapy (which had to be specified) and others (which had to be specified). Finally, physicians participating in the survey were asked to indicate the exact final descriptive diagnosis reported in the medical records at discharge together with classification codes (ICD-09).
2.1. Data collection and statistical analyses
The completed forms were received either by e-mail or fax and analysed at the end of the survey. The NCSS 97 Statistical System for Windows (NCSS, Kaysville, UT) was used for all statistical analyses. Descriptive statistics including mean, ranges and standard deviation values were calculated for all the continuous variables. We used Mann–Whitney U-test to compare continuous data. The χ2-test or Fisher exact test when appropriate was used to compare the categorical data; the results are presented as counts with percentages. All reported p values are based on two-tailed tests of significance. Comparisons were considered significantly different if p
<0.05.
3. Results
A total of 218 forms corresponding to incident CCA cases seen over the 12 months were collected. These forms originated from 41 different centres (mostly tertiary centres) including 5 Surgery units, 1 Oncology unit and 35 Gastroenterology or Internal Medicine units (Appendix A). Nineteen centres were from northern Italy, 8 from southern Italy and 14 from central Italy. Of the 218 incident CCA cases (age=69.01±10.19 years; BMI=24.29±3.60), 115 were males (age 68.00±9.69 years; BMI=24.92±2.99) and 103 females (age 70.04±11.09 years; BMI=23.45±3.93). The majority of the CCA patients lived in central Italy (109 cases) followed by the north (58 cases) and the south (51 cases) Italy.
Based on the current classification [2], 102/218 (46.79%) CCAs were EH-CCA and 116/218 (53.21%) were IH-CCA (Table 1). IH-CCAs were more frequently males (M/F=63/53, ratio=1.2:1) whilst EH-CCAs were similarly distributed by sex (M/F=52/50). The age at diagnosis was significantly higher for EH-CCA than IH-CCA (71.50±10.69 years vs 66.93±9.73 years, p=0.001), the difference being due to females (73.36±10.46 years vs 66.98±10.85 years, p=0.005) but not to males (69.48±10.7 years vs 67.06±8.79 years). BMI at diagnosis was similar between EH-CCA (24.08±3.18) and IH-CCA (24.44±3.84) and this occurred without differences between females. No difference in the geographic residence of patients was found between IH-CCA and EH-CCA (centre=56 vs 53; north=32 vs 26; south=28 vs 23, Table 1).
Table 1. Characteristics of IH-CCA and EH-CCA patients.
| IH-CCA (n=116) | EH-CCA (n=102) | |
|---|---|---|
| Age (years) | 66.93±9.73 (range 36–90) | 71.50±10.69† (range 42–94) |
| Geographic residence | ||
| North Italy | 32 (27.58) | 26 (25.49) |
| Central Italy | 56 (48.27) | 53 (51.96) |
| South Italy | 28 (24.13) | 23 (22.54) |
| Gender, males/females | 63/53 | 52/50 |
| Age, males (years) | 67.06±8.79 (range 40–90) | 69.48±10.70 (range 44–88) |
| Age, females (years) | 66.98±10.85 (range 36–85) | 73.36±10.46* (range 42–94) |
| BMI | 24.44±3.84 (range 14–35) | 24.08±3.18 (range 17–31) |
†p |
*p |
As far as smoking habits were concerned, only 26/218 CCA cases were current smokers (12 IH-CCA vs 14 EH-CCA) whilst 55/218 were past smokers with a predominance (p=0.043) in IH-CCA (36/116, 31%) with respect to EH-CCA (19/102, 18.6%). Alcohol consumption was denied in 115/218 cases (67 IH-CCA vs 48 EH-CCA), with 38 cases continuing to drink alcohol (23 IH-CCA vs 15 EH-CCA, p=ns) and 28 reporting past alcohol use (15 IH-CCA vs 13 EH-CCA). However, 37 CCA cases (11 IH-CCA and 26 EH-CCA) avoided answering the question related to alcohol use.
The pathologies associated with CCA have been reported in Table 2. Liver cirrhosis of any etiology as well as viral cirrhosis (HCV-Ab+ or HBsAg+ cirrhosis together) was more frequently (p<0.05) associated with IH-CCA than with EH-CCA. Individually, however, HCV-Ab+ and HBsAg+ cirrhosis were not statistically differently distributed between IH-CCA and EH-CCA. Positivity for viral hepatitis markers (HCV-Ab+, HBsAg+), independently from the presence or absence of cirrhosis, also predominated in IH-CCA vs EH-CCA (p=0.048) but, when HCV-Ab or HBsAg positivity were considered separately the statistical significance was lost. No significant difference was found between EH-CCA and IH-CCA in the percentage of patients with a past history of cholecystectomy (18.6% vs 14.6%), endoscopic sphincterotomy (3.9% vs 2.6%) or open surgery of the biliary tree (6.9% vs 4.3%), which were shown together in Table 2 (i.e. past surgery biliary tree). PSC was reported only in one EH-CCA patient.
Table 2. Conditions associates with IH-CCA or EH-CCA.
| IH-CCA (n=116) | EH-CCA (n=102) | |
|---|---|---|
| Cirrhosis total | 16 (13.79) | 5† (4.31) |
| Cirrhosis viral | 12 (10.34) | 3* (2.94) |
| Cirrhosis HCV+ | 7 (6.03) | 2 (1.96) |
| Cirrhosis HBsAg+ | 4 (3.44) | 1 (0.98) |
| Cirrhosis HCV+/HBsAg+ | 1 (0.86) | 0 |
| Positive hepatitis virus markers | 35 (30.17) | 19‡ (18.62) |
| HCV+ total | 18 (15.51) | 8 (7.84) |
| HCV+ without cirrhosis | 10 (8.62) | 5 (4.31) |
| HBsAg+ total | 17 (14.65) | 10 (9.80) |
| HBsAg+ without cirrhosis | 12 (10.34) | 8 (7.84) |
| HCV+/HBsAg+ | 0 | 1 (0.98) |
| PSC | 0 | 1 (0.98) |
| Past surgery of the biliary tree (cholecystectomy, sphincterotomy or open surgery of biliary tree) | 25 (21.55) | 30 (29.41) |
*p |
†p |
‡p |
Based on the clinical history, the onset of the disease was estimated to occur earlier for IH-CCA than for EH-CCA (4.35±4.09 months vs 3.07±2.86 months, p=0.009). Symptoms characterizing the clinical presentation of the disease have been reported in Table 3. Painless jaundice without other symptoms represented the clinical presentation of 21.6% EH-CCA whilst this was rare in IH-CCA (2.58%, p<0.01). In general, whilst 74% of EH-CCA presented with jaundice (with or without other symptoms), this occurred only in 28% of IH-CCA cases (p<0.01). Abdominal pain without other symptoms represented a frequent clinical presentation of IH-CCA (19.8%) with respect to EH-CCA (3.9%, p<0.01). When abdominal pain was associated with jaundice, however, this occurred 2-fold more frequently in EH-CCA than IH-CCA (24.5% vs 10.3%, p<0.01). Pruritus alone was referred only in 2 EH-CCA cases but in no case of IH-CCA and, in general, pruritus was rarely encountered both in EH-CCA (6.9%) and in IH-CCA (4.3%). Systemic symptoms (weight loss, fatigue, vomiting, nausea, diarrhoea and malaise) occurred more frequently in IH-CCA than EH-CCA but the difference failed to reach statistical significance. Finally, diagnosis of IH-CCA was obtained in the absence of symptoms (incidental) in 32/116 cases (28%) whilst this rarely occurred in EH-CCA (6 cases, 5.9%; p<0.01).
Table 3. Presenting symptoms.
| IH-CCA (n=116) | EH-CCA (n=102) | |
|---|---|---|
| Jaundice total | 32 (27.58) | 75* (73.52) |
| Painless jaundice alone | 3 (2.58) | 22* (21.56) |
| Painless jaundice+other symptoms (malaise, weight loss, fever, vomiting, diarrhoea) | 17 (16.65) | 28† (27.45) |
| Abdominal pain total | 49 (42.24) | 36 (35.29) |
| Abdominal pain alone | 23 (19.82) | 4* (3.92) |
| Jaundice+abdominal pain | 12 (10.34) | 25* (24.50) |
| Abdominal pain+other symptoms (malaise, weight loss, fever, vomiting, diarrhoea) | 14 (12.06) | 7 (6.8) |
| Weight loss total | 21 (18.10) | 10 (9.8) |
| Weight loss alone | 7 (6.03) | 3 (2.94) |
| Weight loss+other symptoms (abdominal pain, fever, vomiting, malaise, diarrhoea) | 14 (12.06) | 7 (6.8) |
| Pruritus total | 5 (4.31) | 7 (6.86) |
| No symptoms | 32 (27.58) | 6* (5.88) |
†p |
*p |
Table 4 reports the imaging procedures performed during the diagnostic work-up. Apart from 3 IH-CCA cases and 1 EH-CCA case where only abdominal US was performed, all the remaining cases received multiple imaging techniques. Generally, for IH-CCA, 96% of cases had abdominal US, 50% MR+MRCP and 73% CT. For EH-CCA, 80% had abdominal US, 59% had MR+MRCP and 73% CT. ERCP was performed in 49% of EH-CCA cases. In 72/116 IH-CCA (62%) and in 74/102 EH-CCA (73%) at least 3 different imaging procedures were carried out during the diagnostic work-up.
Table 4. Imaging procedures performed in the diagnostic work-up.
| IH-CCA (n=116) | EH-CCA (n=102) | |
|---|---|---|
| Abdominal US alone | 3 | 1 |
| Abdominal US+others | 107 | 82 |
| MR+MRCP+others | 58 | 60 |
| CT+others | 85 | 77 |
| US+MR+MRCP | 14 | 10 |
| US+CT | 13 | 5 |
| PET+others | 10 | 3 |
| Endo-US+others | 2 | 11 |
| ERCP+others | 3 | 49 |
| 2 imaging procedures | 41 | 26 |
| 3 imaging procedures | 53 | 38 |
| 4 imaging procedures | 19 | 36 |
Table 5 reports the anatomical localisation of CCA defined on the basis of the imaging studies and/or surgical description. For IH-CCA, 91/116 (78.44%) presented as a single hepatic mass, 10/116 as double nodules and the remaining 15 cases as multifocal disease (>2 nodules). For the IH-CCA presenting as single mass, the IV liver segment was involved in the majority of cases, being the only segment involved in 24 cases and being interested together with other segments in additional 24 cases. As far as the EH-CCA was concerned, by excluding 12 unspecified cases, 51/91 (56.04%) presented as peri-hilar (Klatskin) tumour, 25 (27.47%) as cancer involving the distal bile duct and 15 (16.48%) involving the middle common bile duct.
Table 5. Anatomical localisation of IH-CCA and EH-CCA.
| IH-CCA, n=116 | N (%) | EH-CCA (n=102) | N (%) |
|---|---|---|---|
| Single mass | 91 (78.44) | Hilar (Kaltskin) total | 51 (50) |
| 2 nodules | 10 (8.62) | Hilar, Bismuth I | 15 |
| >2 nodules | 15 (12.93) | Hilar, Bismuth II | 12 |
| Location of single mass | Hilar, Bismuth III | 15 | |
| Single mass involving only one liver segment | Hilar, Bismuth IV | 9 | |
| Left lobe (segment II, III or IV) | 28a | Middle | 15 (14.70) |
| Right lobe (segment V, VI, VII or VIII) | 11 | Distal | 25 (24.50) |
| Single mass involving more liver segments | 52b | Unspecified | 12 |
a24/28 involved only the IV segment. |
b24/52 involved also the IV segment. |
Table 6 reports how tissue-proven (histology, cytology) diagnosis of CCA was obtained. For IH-CCA, histological diagnosis was obtained exclusively by US-guided biopsy in 19.8% of cases, only by surgery in 27.6% and by both US-guide biopsy+surgery in 31% of cases. In the remaining 21.5% of the IH-CCA cases, no histological confirmation was obtained and therefore definitive diagnosis was based exclusively on clinical, imaging and/or biomarkers. For EH-CCA histological or cytological diagnosis was obtained at ERCP (biopsy, brushing or bile cytology) in 30% of the cases, at surgery in 19.6% and at both ERCP+surgery in 12% of the cases. In 19.6% of EH-CCA diagnosis was based only on imaging results. Overall, a pathologically confirmed diagnosis was obtained in 78.45% of IH-CCA and in 80.40% of EH-CCA.
Table 6. Tissue-proven diagnosis of IH-CCA and EH-CCA.
| IH-CCA (n=116) | EH-CCA (n=102) | |
|---|---|---|
| US-guided percutaneous bx. alone | 23 (19.82) | 7 (6.86) |
| Endoscopic biopsy alone | 18 (17.64) | |
| Brushinga alone | 11 (10.78) | |
| Citologya alone | 2 (1.96) | |
| Surgery alone | 32 (27.58) | 20 (19.60) |
| US-guided percutaneous biopsy+surgery | 36 (31.03) | 2 (1.96) |
| Endoscopic biopsya+surgery | 10 (9.80) | |
| Brushinga+surgery | 2 (1.96) | |
| Citologya+surgery | 2 (1.96) | |
| Brushinga+endoscopic bx.a | 1 (0.98) | |
| Brushinga+cytologya | 3 (1.66) | |
| Endoscopic biopsya+cytologya | 2 (1.96) | |
| Brushinga+endoscopic biopsya+cytologya | 0 | |
| Brushinga+endoscopic biopsya+surgery | 2 (1.96) | |
| Not obtained | 25 (21.55) | 20 (19.60) |
aPerformed at ERCP. |
Table 7 reports the procedures used to stage the disease. Staging was mostly based on imaging procedures but, PET was rarely (6 cases) used in IH-CCA and never in EH-CCA. Laparotomy and/or laparoscopy were rarely used for staging purposes. 23/116 IH-CCA vs 25/102 EH-CCAs were reported with distant metastasis.
Table 7. Procedures used for cholangiocarcinoma staging.
| IH-CCA (n=116) | EH-CCA (n=102) | |
|---|---|---|
| MR+MRCP only | 13 | 12 |
| CT only | 34 | 38 |
| Laparoscopy only | 3 | 2 |
| Laparotomy only | 2 | 1 |
| CT+MR+MRCP | 16 | 20 |
| CT+PET | 6 | 0 |
| CT+laparoscopy | 0 | 2 |
| CT+laparotomy | 4 | 2 |
| MR+MRCP+laparoscopy | 2 | 1 |
| MR+MRCP+laparotomy | 1 | 0 |
| Laparoscopy total | 15 | 9 |
| Laparotomy total | 7 | 5 |
| Not reported | 13 | 10 |
As far as treatment was concerned (Table 8), 57% (66/116) of IH-CCA underwent open surgery which was considered curative in 52 out of 66 cases, and palliative in 14 out of 66 cases. IH-CCAs were treated with chemotherapy alone in 10.3% (12/116), with chemotherapy plus bile drainage in 21.5% (14/116) and exclusively with palliative bile drainage in 13.8% (16/116). Finally, 18.1% (21/116) of IH-CCA received no treatment. For EH-CCA, 42% (43/102) underwent open surgery where, in 30 cases surgery was considered curative and in 13 cases palliative. 13.7% (14/102) of EH-CCAs were submitted to bile drainage+chemotherapy or radiotherapy and 38% (39/102) only received palliative bile drainage. Only one patient received exclusively radiotherapy and one chemotherapy+radiotherapy. None of the IH-CCA or EH-CCA patients received photodynamic therapy. Finally, 3.9% (4/102) of EH-CCA received no treatment. Therefore curative surgery was possible in 45% of all IH-CCA and in 29% of all EH-CCA (p=0.025).
Table 8. Treatment modalities.
| IH-CCA (n=116) | EH-CCA (n=102) | |
|---|---|---|
| Surgery only | 57 (49.13) | 19 (18.62) |
| Chemotherapy only | 9 (7.76) | 0 |
| Bile drainage only | 16 (13.79) | 39 (38.23) |
| Surgery+bile drainage | 3 (2.58) | 15 (14.70) |
| Surgery+other treatments (chemotherapy, radiotherapy) | 6 (5.17) | 9 (8.82) |
| Chemotherapy+bile drainage | 3 (2.58) | 6 (5.88) |
| Bile drainage+radiotherapy | 1 (0.86) | 8 (78.43) |
| No treatment | 21 (18.10) | 4 (3.92) |
The classification codes (ICD-09) reported in the medical records at discharge were coherent with the descriptive diagnosis and the information (imaging, surgery, etc.) reported in the form in almost all CCA cases. However, in 5/51 hilar CCA, the code corresponding to intra-hepatic CCA was reported confirming that a misclassification of hilar CCA as intra-hepatic CCA may occur as previously reported [10].
4. Discussion
CCA is commonly considered a rare cancer. However, if we consider the hepatobiliary system as an entire entity, cancers of the gallbladder, intra-hepatic and extra-hepatic biliary tree altogether represent approximately 30% of the total with incidence rates approaching that of hepatocellular carcinoma which is considered the third most common cause of cancer-related death worldwide [8], [9], [10], [11], [12], [13], [14], [15]. In addition, CCA is characterised by a very poor prognosis, with virtually no response to chemotherapeutics and, radical surgery, the only effective treatment, is not frequently applicable due to late diagnosis [1], [2], [3], [4], [5]. Biomarkers for screening programmes and for follow-up of categories at risk are not currently available [16], [17]. Thus, cancers of the biliary tree system should merit much more scientific attention also because a progressive increase in incidence and mortality for this cancer has been reported worldwide [7], [9], [10], [11], [12], [15].
Very few studies have been aimed at evaluating the CCA clinical characteristics, the majority of the reports having origin from the experience of single centres [18], [19], [20]. In this study, we presented the result of an Italian survey on CCA clinical features. After excluding gallbladder cancer, 218 consecutive CCA cases were collected in a 12-month period of time. Recent epidemiological data in Italy [15], originating from the Italian National Institute of Statistics (mortality) and Italian Cancer Registries (incidence) indicated that, in the year 2002, all bile duct cancers (gallbladder excluded) accounted for a total of 1258 deaths (real global data) and for 1872 incident cases as estimated from 13 Italian Cancer Registries. By considering these epidemiologic data [15], in our survey we probably analysed from 12 to 17% of total new CCA cases in Italy. The point meriting careful attention is whether CCA patients analysed in our survey are representative of the total Italian sample. Most of the CCA patients considered in our study were from tertiary centres of Gastroenterology and Internal Medicine and, therefore, they should represent patients needing management of diagnostic work-up, staging or treatment decisions. We probably missed very advanced cases for whom no treatment was possible or directly referred to oncology units for treatment and, in addition, cases with a precise surgical indication which were directly referred to surgeons bypassing our units. Therefore, our findings should be mainly representative of CCA cases referred to units of Gastroenterology and Internal Medicine.
The 218 incident CCA cases analysed in the survey were equally distributed by sex and were, on average, 70 years old confirming that CCA is a cancer of the 6th to 8th decades. Half of the collected cases lived in central Italy, the remaining being equally distributed between northern and southern Italy. IH-CCA and EH-CCA were equally represented in our population. This could be considered a selection bias due to the fact that EH-CCAs are more frequently referred to surgeons because of the obstructive jaundice. However, a number of recent studies documented that the worldwide incidence and mortality for IH-CCA is progressively increasing and approaching that of EH-CCA, which instead is stable or slightly decreasing [8], [9], [10], [11], [12], [13], [14], [15]. This is also true for Italy where, in the year 2003, mortality for IH-CCA surpassed EH-CCA [15]. At diagnosis, EH-CCA patients were older than IH-CCA and, the onset of the disease, based on clinical history, was estimated to occur earlier for IH-CCA than EH-CCA. This could be related to the most frequent clinical presentation of EH-CCA, i.e. jaundice (see above), which induces physicians to prescribe immediate hospitalisation. As far as risk factors were concerned [1], [2], [3], [4], [5], [6], [7], [21] a history of past smoking was more frequent in IH-CCA than EH-CCA whilst, no difference in alcohol consumption was observed. Most importantly, however, liver cirrhosis or viral cirrhosis was much more frequently associated with IH-CCA. Different recent studies have highlighted the role of HCV and HBV as risk factors for CCA and, indeed, viral genome fragments have been demonstrated in CCA cells [22], [23], [24], [25]. Our study seems to confirm the role of hepatitis viruses for IH-CCA where, up to 30% of cases were interested (HCV or HBsAg positive IH-CCA groups together), this being markedly higher than the general Italian population of similar age. In our findings, however, cirrhosis (mostly viral) seems to have a major impact on IH-CCA than hepatitis virus per se. On the other hand, cirrhosis is associated with activation of the resident stem cell compartment located in canals of Hering/bile ductules and featuring markers of cholangiocytes [6], [7]. The origin of CCA from transformed resident stem cells is a currently investigated hypothesis [6], [7] that obviously applies for IH-CCA but not for EH-CCA. Other putative risk factors such as PSC or open surgery of the biliary tree were rarely encountered in our survey and biliary parasitic infections are not common at this latitude. Only one case of PSC was reported in keeping with the low incidence of this disease in our nation. Finally, no putative risk factor for CCA [1], [2], [3], [4], [5], [6], [7] was found in at least 60% of our CCA cases, suggesting that much more research on identifying new risk factors is needed.
As far as clinical presentation was concerned, the first finding is that a totally asymptomatic presentation occurs in 28% of IH-CCA cases where diagnostic work-up started as a consequence of incidental findings at imaging procedures. Asymptomatic presentation is rare in EH-CCA (only 4%) which, instead, presented jaundiced in 74% of cases with (22%) or without abdominal pain (25%), or associated with other symptoms (27%, malaise, weight loss, fever, pruritus or vomiting). For IH-CCA, the clinical presentation is more heterogeneous with abdominal pain or weight loss, alone or associated with general symptoms, representing common presentations. Interestingly, abdominal pain associated with jaundice is more frequently suggestive of EH-CCA whilst when it was reported alone suggested IH-CCA. Our findings concerning the clinical presentation could explain why the clinical onset of the disease before diagnosis is shorter for EH-CCA where jaundice induces a more rapid hospitalisation with respect to IH-CCA where management of abdominal pain could be performed as an out-patient.
As far as the morphological characteristics and the anatomical localisation is concerned, it is of interest that approximately 80% of IH-CCA presented as a single mass mostly located at the IV liver segment, alone or involving contiguous segments. For the EH-CCA, the most frequent form (68%) was the Klatskin tumour.
Management of CCA diagnosis is time- and cost-consuming since 70% of our CCA patients have performed more than 3 imaging procedures. Although current recommendations or guidelines [1], [2], [3], [4], [5], [26], [27] indicate MR+MRCP as the preferred imaging procedure to be performed after abdominal US, for the suspicion of CCA, CT continues to be used, at least in Italy, more frequently than MR. In clinical practice, it is common to diagnose CCA on the basis of clinical/laboratory/imaging studies without tissue-proven evidence of tumour. However, this occurred in only 20% of the cases in our survey. As expected, the definitive diagnosis is based mostly on US-guided bx. for IH-CCA and on ERCP for EH-CCA. As far as the staging procedure was concerned, it is interesting that PET was rarely used in our hospital units and this was in spite of recent reports or recommendations [1], [2], [3], [4], [5], [26], [27], [28] suggesting the superiority of this procedure with respect to CT or MRI for CCA staging.
Classification of CCA by the ICD-09 codes was correctly performed in almost all the cases and this is important for the epidemiological studies based on the registry data. However, in 5/51 hilar CCA, the cancer was misclassified as intra-hepatic CCA and this re-presents the problem previously discussed by Welzel et al. [10].
As far as treatment is concerned, 57% of IH-CCA vs 42% of EH-CCA underwent open surgery. However, surgery was considered curative in 45% IH-CCA and 29% EH-CCA. This demonstrates that the only effective treatment so far available (i.e. surgery) is applied more frequently in IH-CCA than EH-CCA. Despite different recent reports suggested the benefit of photodynamic and radiotherapies, these treatments were rarely used in our hospital units.
The current investigation has some limitations since data have been obtained through a preformed questionnaire and most patients were followed by tertiary centres. With these limitations, this is the first large multicentre study describing the clinical characteristics of patients with IH-CCA or EH-CCA.
In conclusion, in Italy IH-CCA is diagnosed as frequently as EH-CCA. In comparison to EH-CCA, IH-CCA occurred at a younger age and is more frequently associated with smoking habits and cirrhosis. In 1/5 cases, IH-CCA diagnosis was made incidentally without symptoms whilst most EH-CCAs were jaundiced at the diagnosis (1/5 painless). Multiple procedures were performed in the clinical work-up confirming that CCA management is time- and cost-consuming. Finally, curative surgical treatment is more frequently performed for IH-CCA than EH-CCA.
Conflict of interest statement
List of abbreviations
CCA, cholangiocarcinoma; IH-CCA, intra-hepatic cholangiocarcinoma; EH-CCA, extra-hepatic cholangiocarcinoma; BMI, body mass index; PSC, primitive sclerosing cholangitis; US, abdominal ultrasonography; MR, magnetic resonance; MRCP, MR-cholangiopancreatography; CT, computed tomography; PET, positron emission tomography; EUS, endoscopic US; ERCP, endoscopic retrograde cholangiopancreatography.
Acknowledgements
We thank T. Dornbusch for English Editing, A. Vestri for assistance in statistical analysis and L. Capocaccia for constructive criticisms.
Appendix A. Survey participants with location of the hospital unit in parenthesis
M.C. Bragazzi, A.F. Attili (Rome); F. Nudo, P.B. Berloco (Rome); Angeletti (Rome); C. Napoli, D. Alvaro (Latina); E. Piras (Cagliari); M. Marzioni, G. Fava, A. Benedetti (Ancona); L. Petraccia, A. Fraioli, S. Brozzetti, A. Tocchi (Rome); B. Coco, M.R. Brunetto (Pisa); A.M. De Rose, G. Nuzzo (Rome); G. Del Vecchio, A. Caruso, F. Pallone (Rome); P. Serra (Rome); I. Deli, G. Delle Fave (Rome); E. Villa (Modena); V. Cesario, A. Gasbarrini (Rome); A. Borghi, F. Piscaglia, L. Bolondi (Bologna); G. Tonini (Rome); G. Fornaciari (Reggio Emilia); S. Tremosini (Modena); M.C. Tesini, C. Vandelli (Modena); V. Scuderi, F. Calise (Naple); V.O. Palmieri, G. Palasciano (Bari); A. Rianda, N. Alessi, A. Craxì (Palermo); C. Saitta, G. Raimondo (Messina); G. Montalto (Palermo); G. Scaglione (Benevento); R. Carratu’ (Naple); G. Falzone (Siderno); M. Camassa (Milan); P. Bertoli (Milan); C. Bertelli, A. Fracanzani, S. Fargion (Milan); L. Caliari, P. Invernizzi, M. Podda (Milan); F. Salerno (Milan); S. Saibeni, S. Bruno (Milan); A. Salmi (Brescia); E.G. Giannini (Genova); G. Casella, V. Baldini (Desio); R. Ciccocioppo (Pavia); M.G. Marinone, Fornari (Piacenza); G. Marin (Verona), L. Frevola, M. Colombo (Milan); R. Cannizzaro (Aviano).
References
- . Cholangiocarcinoma: advances in pathogenesis, diagnosis, and treatment. Hepatology. 2008;48:308–321
- . Cholangiocarcinoma. Gastroenterology. 2005;128:1655–1667
- Cholangiocarcinoma and its management. Gut. 2007;56:1755–1756
- . Cholangiocarcinoma: emerging approaches to a challenging cancer. Curr Opin Gastroenterol. 2007;23:317–323
- . Cholangiocarcinoma: natural history, treatment, and strategies for surveillance in high-risk patients. J Clin Gastroenterol. 2008;42:178–190
- . New insights on the molecular and cell biology of human cholangiopathies. Mol Aspects Med. 2008;29:50–57
- Proliferating cholangiocytes: a neuroendocrine compartment in the diseased liver. Gastroenterology. 2007;132:415–431
- . Increasing incidence and mortality of primary intrahepatic cholangiocarcinoma in the United States. Hepatology. 2001;33:1353–1357
- . A comparison of trends in the incidence of hepatocellular carcinoma and intrahepatic cholangiocarcinoma in the United States. Cancer Epidemiol Biomarkers Prev. 2006;15:1198–1203
- Impact of classification of hilar cholangiocarcinomas (Klatskin tumors) on the incidence of intra- and extrahepatic cholangiocarcinoma in the United States. J Natl Cancer Inst. 2006;98:873–875
- Rising incidence of intrahepatic cholangiocarcinoma in the United States: a true increase?. J Hepatol. 2004;40:472–477
- Do increases in mortality from intrahepatic cholangiocarcinoma reflect a genuine increase in risk? Insights from cancer registry data in Scotland. Eur J Cancer. 2003;39:2087–2092
- . Worldwide trends in mortality from biliary tract malignancies. BMC Cancer. 2002;2:10
- Trends in the incidence of primary liver and biliary tract cancers in England and Wales 1971–2001. Br J Cancer. 2006;94:1751–1758
- Alvaro D, Crocetti E, Ferretti S, et al. Descriptive epidemiology of cholangiocarcinoma in Italy. Digest Liver Dis; in press.
- . Serum and bile biomarkers for cholangiocarcinoma. Curr Opin Gastroenterol. 2009;25:279–284
- Serum and biliary insulin-like growth factor I and vascular endothelial growth factor in determining the cause of obstructive cholestasis. Ann Intern Med. 2007;147:451–459
- Cholangiocarcinoma: a 7-year experience at a single center in Greece. World J Gastroenterol. 2008;14:6213–6217
- Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg. 1996;224:463–473
- . Incidence, etiologic aspects and clinicopathologic features in intrahepatic cholangiocellular carcinoma—a study of 51 cases from a low-endemicity area. Acta Oncol. 1998;37:77–83
- Cholangiocarcinoma in primary sclerosing cholangitis: risk factors and clinical presentation. Scand J Gastroenterol. 2002;37:1205–1211
- Risk of hepatobiliary and pancreatic cancers after hepatitis C virus infection: a population-based study of U.S. veterans. Hepatology. 2009;49:116–123
- Hepatitis B virus infection and intrahepatic cholangiocarcinoma in Korea: a case–control study. Am J Gastroenterol. 2008;103:1716–1720
- Intrahepatic cholangiocarcinoma and hepatitis C and B virus infection, alcohol intake, and hepatolithiasis: a case–control study in Italy. Cancer Causes Control. 2001;12:959–964
- Incidence of primary cholangiocellular carcinoma of the liver in Japanese patients with hepatitis C virus-related cirrhosis. Cancer. 2000;88:2471–2477
- . Cholangiocarcinoma: current concepts and insights. Hepatology. 2003;37:961–969
- Guidelines for the diagnosis and treatment of cholangiocarcinoma: consensus document. Gut. 2002;51(Suppl. 6):VI1–VI9
- 18F-fluorodeoxyglucose positron emission tomography influences management decisions in patients with biliary cancer. J Am Coll Surg. 2008;206:57–65
PII: S1590-8658(10)00166-0
doi:10.1016/j.dld.2010.05.002
© 2010 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Inc All rights reserved.
