Article Text


Small fry starting to make waves
  1. Department of Histopathology, ICSM at Hammersmith Campus
  2. Du Cane Road, London W12 0NN, UK
  3. m.alison{at}

Statistics from

See article on page 816

The liver, perhaps because of its crucial role in metabolism and detoxification of many potentially hazardous xenobiotics, has evolved not one but three stem cell populations. The one that usually operates is the hepatocyte population itself. Hepatocytes were formerly (and incorrectly) considered to have a limited division potential because only 2–3 rounds of cell division occurred after a two thirds partial hepatectomy but this was all that was required to restore liver mass! However, their full division potential has been unmasked through the study of models of hepatocyte transplantation, and they fully deserve the appellation of “functional stem cells” with at least some of them being capable of in excess of 100 divisions. A second population comes into play when either hepatocyte regeneration is compromised after injury or when parenchymal damage is particularly severe. If we did not already know it, rats and humans are different and this extends to the intrahepatic biliary tree. In rats, the terminal bile ductules (canals of Hering) barely extend beyond the portal tract, and the so-called ductular oval cell response that occurs after parenchymal injury when hepatocyte regeneration is inhibited emanates from the intraportal bile ducts and canals of Hering.1 On the other hand, in humans, the canals of Hering extend into the proximate third of the hepatic lobule, and the ductular reactions that are seen after massive necrosis are considered to be the proliferative reactions of these extended canals of Hering,2 being akin to a trip wire that “senses” severe injury. Thus in both rats and humans, the cholangiocyte population is considered to harbour a facultative stem cell population that acts as a “back up” to the functional stem cells. More recently, a third liver stem cell candidate has become apparent, namely bone marrow stem cells that appear to be able to differentiate into hepatocytes and cholangiocytes,3 4 and moreover are capable of making a substantial contribution to liver repopulation in a murine model of metabolic liver disease.5

The now classical experiments on epidermal carcinogenesis indicated that the target cell for genotoxic chemicals is likely to be a long lived cell (that is, a stem cell) and indeed popular opinion opines that neoplasia, like metaplasia, is a stem cell disease. So what is the carcinogen target cell for the development of primary tumours of the liver? As itemised above, there are at least three possible different cellular targets,6 although for tumours of the biliary epithelia it is likely to be directly from a cholangiocyte, but that cell could have been bone marrow derived!7 In 1998 in the UK, primary tumours of the liver accounted for only <1% of the total cancer burden and for 1% of deaths from cancer ( However, in this issue of Gut, Simon Taylor-Robinson and colleagues flag up a seemingly inexorable rise in the number of deaths in England and Wales from intrahepatic cholangiocarcinoma,8 a relatively rare cancer with a dismal prognosis (see page 816). They report a 15-fold increase in age specific mortality rates in the 45 and over age groups over a 30 year period, resulting in this malignancy now being the commonest primary liver tumour and responsible for more deaths than hepatocellular carcinoma in England and Wales. In China, Japan, and South East Asia, infection by liver flukes such as Clonorchis sinensis and Opisthorchis viverriniis endemic, and here the chronic biliary irritation associated with these food borne trematodes is a predisposing factor: in areas of Thailand, with the highest incidence of O viverrini infection, the incidence of cholangiocarcinoma is 40 times the highest incidence outside Thailand and cases of cholangiocarcinoma outnumber cases of hepatocellular carcinoma.

So what is the aetiology of intrahepatic cholangiocarcinoma in England and Wales, and is increased proliferation of the putative target cell (cholangiocyte) a significant risk factor? Primary sclerosing cholangitis, an inflammatory disease of bile ducts, is certainly a predisposing factor in a minority of cases, and in models of chemically induced hepatocellular carcinoma many established hepatocarcinogens are only effective if coupled to enhanced hepatocyte turnover in an otherwise essentially mitotically quiescent organ. An eminent hepatopathologist Ken Weinbren once remarked that “bile ducts proliferate at the drop of a hat” and apart from their function as facultative stem cells, cholangiocyte proliferation can be stimulated by the likes of bile acids and oestrogens,9 and indeed many compounds with oestrogenic properties can probably enter the food chain. As to the identity of the carcinogenic agent(s), we are largely in the dark; polychlorinated biphenyls are ubiquitous and persistent environmental contaminants whose soil concentrations have been rising for some time, even before the rises in intrahepatic cholangiocarcinoma reported here. Rated as “probable human carcinogens” by the IARC, they may be linked to breast cancer but an association with liver tumours has yet to be proved. The genetic alterations implicated in the development and progression of intrahepatic cholangiocarcinoma are often the usual culprits such asTP53,mdm2, K-ras, andINK4a, although a dysfunctional p53 pathway throughTP53 mutation and/or Mdm2 overexpression seems to be the most consistent reported abnormality.10Interestingly, the types of mutations in theTP53 gene in European patients are dissimilar from those in East Asian patients suggesting different environmental factors are at work.11

Although there are widely acknowledged increases in the levels of malignant melanoma and prostate cancer, for example, much bigger hitters in the league of cancer statistics, the report by Taylor-Robinson et al seems to highlight a much more striking increase in a tumour still relatively uncommon in the UK. This increase appears “real” and could not reasonably be attributed to “diagnosis transfer” or the widespread adoption of improved investigative procedures. Clearly the search for carcinogenic agents in bile should be a high priority.

See article on page 816


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles