Introduction

1. An increase in the age-standardised mortality rate (ASMR) for all causes of liver cancer has been documented in England and Wales over the period 1979-1998. A preliminary investigation suggested that the increase in mortality from liver cancer may in part be due to an increase in mortality from intrahepatic cholangiocarcinoma.(1) Intrahepatic cholangiocarcinoma is a relatively rare tumour in the UK and the prognosis for patients with this tumour is poor.(2) It is therefore necessary to review any evidence for an increase in the incidence of this tumour in order to establish whether any documented increase in mortality is real or artefactual. For example, changes in classification of tumour type, changes in the collection or coding of mortality data, or the introduction of improved diagnostic procedures for cholangiocarcinoma, such as endoscopic retrograde cholangiopancreatography (ERCP), could account for the observed increase in incidence of this tumour. However, should a genuine increase in the mortality from intrahepatic cholangiocarcinoma in England and be established, it would be important to consider the aetiological factors that might be responsible for such an increase and any preventative action that could be taken.

Background to COC consideration

Evidence for an increase in mortality rate from intrahepatic cholangiocarcinoma in England and Wales 1968-1998.

2. In November 1999, the Committee reviewed a prepublication copy of a draft paper by Taylor-Robinson et al which presented a detailed analysis of mortality data from 1968 to 1986 for tumours of the liver, and of the pancreas, obtained from the Office of National Statistics. Members heard a presentation of these data from Professor Howard Thomas and colleagues at the Division of Medicine, Imperial College of Science, Technology and Medicine. The total number of deaths attributed to particular tumours (using the International Classification of Disease (ICD) 8th and 9th revisions) were analysed by year and sex. The tumours considered in the report were:

3. ASMRs per 100,000 population and, for intrahepatic cholangiocarcinoma, age-specific mortality rates (AspMR) per 100,000 population (using four age bands: 20-44y, 45-64y, 65-74y and 75y+) were calculated. Mortality data were considered to be a general indicator of incidence because prognosis from cancer in the liver is poor. Detailed evaluation of the data on intrahepatic cholangiocarcinoma was undertaken to ascertain whether the observed increase in mortality from this tumour represented a true increase in the numbers of this rare form of cancer.

4. The authors noted that in 1978 there was a total of 95 deaths in England and Wales reported from intrahepatic cholangiocarcinoma whereas there were 736 cancer deaths attributed to this tumour in 1996. The increase in ASMR over the period 1968 to 1996 was from 0.1/100,000 to 1.22/100,000 in males and from 0.05/100,000 to 0.92/100,000 in females.

5. In June 2001 the COC saw a revised version of the paper by Taylor-Robinson et al(3) (and a subsequent correction)(4) which included additional data on mortality from intrahepatic carcinoma for 1997-1998. ASMR per 100,000 population increased to 1.37 and 1.12 in 1998 for males and females, respectively, bringing the total number of deaths for that year to 864. AspMR per 100,000 population increased from 1968 to 1996, on average, 12-fold in ages 45 and over, with larger increases in older ages and women.

6. In 2001, the Committee also saw two publications reporting increases in the reported incidence and mortality rates for intrahepatic cholangiocarcinoma in the US.(5,6) The ASMR per 100,000 population in the US was reported to have increased from 0.07 in 1973 to 0.69 in 1997; and the estimated incidence rates per 100,000 from 0.13 in 1973 to 0.67 to 1997.(6)

COC consideration

7. The Committee observed that the exceptionally large increase in ASMR for intrahepatic cholangiocarcinoma documented in the paper by Taylor-Robinson et al (i.e. 13.7 fold in males, 22.4 fold in females over the period 1968-1998) was a highly unusual finding. The Committee noted that a number of potential confounding factors had been considered by the authors, such as changes in criteria for ICDisease codes (ICD revisions 8 and 9), the introduction of ERCP as a new diagnostic technique for facilitating precision in the location of site for several cancers of the hepatobiliary system in the mid to late 1970s, and the potential for diagnostic transfer from tumours of the pancreas, gallbladder and extrahepatic biliary tree. The Committee commented that the reported increase in mortality from intrahepatic cholangiocarcinoma appeared to coincide with the introduction of ERCP as a new diagnostic technique. Members considered that the interpretation of the apparent increasing trend in ASMR for intrahepatic cholangiocarcinoma was complicated in that the increase extended beyond the period required for learning and correctly applying this diagnostic technique. However, Members considered that a small diagnostic transfer from secondary pancreatic cancer to intrahepatic cholangiocarcinoma, occurring as a result of increasing use of new diagnostic techniques, could account for the observed results either totally or in part. Members also considered that diagnostic transfer from secondary adenocarcinoma in the liver, which is relatively common and may be indistinguishable from intrahepatic cholangiocarcinoma histologically, might be another explanation. Members also noted that, in both England and Wales and in the US, there had been a steady decrease in reported mortality rates from extrahepatic biliary system tumours as the reported mortality rates of intrahepatic cholangiocarcinoma had risen and considered that diagnostic transfer from extrahepatic to intrahepatic biliary system tumours could account, in part, for the reported increase in the latter. The Committee considered that the ASpMR data, which indicated higher increases in the rate of intrahepatic cholangiocarcinoma in older age groups, supported the hypothesis of diagnostic transfer. It was to be expected that cancer would be more common in older patients. In the past, investigation of cancer in the liver tended to be less extensive in older than younger patients, which may have led to less precise diagnosis and the highest level of underreporting of intrahepatic cholangiocarcinoma in older age groups. Changes in clinical practice had made it increasingly more likely that older patients would receive a precise diagnosis and thus it was in these age groups that the greatest rate of increase in recorded cases of intrahepatic cholangiocarcinoma might be anticipated.

8. The Committee noted that little was known about the aetiology of intrahepatic cholangiocarcinoma. Tumours of the intrahepatic bile ducts are common in South East Asia but are relatively rare in Western populations.7 The most recent information from the Office of National Statistics confirms this. In 1998 there were 1921 deaths from liver cancer in England and Wales (i.e. 1.41% as a proportion of deaths from all malignant tumours) and 86 deaths from intrahepatic cholangiocarcinoma (i.e. 0.63% as a proportion of deaths from all malignant tumours). The causes of intrahepatic cholangiocarcinoma have not been fully elucidated and there are only three known risk factors. Infestation by liver flukes (distomes Clonorchis sinensis and Opisthorchis viverrini) is associated with these tumours and geographical distribution of intrahepatic cholangiocarcinoma coincides with endemic areas for infestation with such parasites. Primary sclerosing cholangitis is also associated with an increased risk of intrahepatic biliary tract carcinoma but this is likely to account for only a small number of the total cases of intrahepatic cholangiocarcinoma. Thorium dioxide (Thorotrast ), which was used in the 1930-1940s as a radiological contrast agent in medicine, is also known to induce cholangiocarcinoma in humans. (7,8)

COC Conclusions

9. The Committee noted that a substantial increase in the reported rate of mortality from intrahepatic cholangiocarcinoma had been documented in both England and Wales and in the US over the last 30 years or so and concluded that changes in diagnostic standards over time could contribute to this increase. It was therefore important to undertake additional investigations before a definite conclusion could be reached about whether there had been real increase in the incidence of this tumour. The Committee recommended that:

i) an integrated pathological and clinical review of cases was needed to ascertain the accuracy of diagnosis of intrahepatic cholangiocarcinoma and, in particular, the potential for diagnostic transfer from secondary adenocarcinoma in the liver, for example, from carcinoma of the pancreas to intrahepatic cholangiocarcinoma.

ii) further evaluation of time-trends for intrahepatic cholangiocarcinoma in other countries, particularly with reference to the introduction of improvements in diagnostic imaging, would be helpful.

iii) consideration should be given to examination of information on the diagnosis of intrahepatic cholangiocarcinoma in a number of specialist liver units before and after the introduction of better diagnostic imaging techniques.

iv) whether or not the recorded increase in incidence of intrahepatic
cholangiocarcinoma was artefactual, it was clearly higher than was previously believed to be the case. Given the poor prognosis from this cancer, it was important to improve our understanding of its aetiology

v) the topic should be kept under review.

March 2002
References

1. Taylor-Robinson SD, Foster GR, Arova S, Hargreaves S and Thomas HC (1997). Increase in primary liver cancer in the UK, 1979-1994. Lancet, 340, 1142-1143.

2. Gennari L, Doci R, and Bozzetti F (1995). Liver Tumours. Chapter 7, pp1201-1220 in Oxford Textbook of Oncology, Volume 1, edited Peckham M, Pinedo HM and Keronesi U. Oxford University Press, Oxford.

3. Taylor-Robinson SD, Toledano MB, Arora S, Keegan TJ, Hargreaves S, Beck A, Khan SA, Elliott P and Thomas HC (2001). Increase in mortality rates from intrahepatic cholangiocarcinoma in England and Wales 1968-1998. Gut, 48, 816-820.

4. Corrections. Gut, 49, 315 (2001).

5. Gumstrop B, Guruprasad P Aithal. Striking trends in the cholangiocarcinoma
incidence and mortality in the United States of America 1979-1998. Abstract presented at the American Gastroenterology Association Meeting, Atlanta, May 2001.

6. Patel Tushar (2001). Increasing incidence and mortality of primary intrahepatic
cholangiocarcinoma in the United States. Hepatology, 33(6), 1353-1357.

7. IARC (1990). Cancer: Causes Occurrence and Control . Edited Tomatis L,
IARC Scientific Publications No 100, Lyon, France.

8. Lyon-Murray I, M (1996). Liver Tumours. Section 14.32 in Chapter 14 on Gastroenterology in Oxford Textbook of Medicine, third edition, edited Weatherall DJ, Ledingham JGG, and Warell DA. Oxford Scientific Press, Oxford.

Top

copyright: © | updated: 29 July, 2002