Rising trends in cholangiocarcinoma: is the ICD classification system misleading us?

Background & aims: Cholangiocarcinomas (CC) can be sub-divided into intrahepatic (IHCC) or extrahepatic (EHCC). Hilar or 'Klatskin' tumours are anatomically extrahepatic. Most international studies, also from the UK, report increasing IHCC and decreasing EHCC incidence. The second edition of the International Classification of Diseases for Oncology (ICD-O-2) assigned 'Klatskin' tumours a unique histology code (8162/3), but this was cross-referenced to the topography code for intrahepatic (IHBD) rather than extrahepatic bile duct tumours (EHBD). Under the third ICD-O edition, 'Klatskin' tumours are cross-referenced to either IHBD or EHBD. New editions of the ICD-O classification are adopted at different time points by different countries. We investigated the impact of changing ICD-O classifications and the potential misclassification of hilar/'Klatskin' tumours on bile duct tumour and CC incidence rates in England and Wales and the US. We also examined whether coding practices by cancer registries in England and Wales could be influencing these rates.

Methods: We analysed age-standardised incidence rates (ASIR) in England and Wales for IHBD and EHBD tumours between 1990 and 2008, then transferred all 'Klatskin' tumours from IHBD to EHBD and reanalysed rates from 1995, when ICD-O-2 was introduced in the UK. We also compared trends in IHBD, EHBD, and 'Klatskin' tumours in England and Wales with those in the USSEER (Surveillance, Epidemiology and End Results) database. Coding practice at Cancer registry level in England and Wales was investigated via a questionnaire completed by all national cancer registries.

Results: In England and Wales, 1990-2008, ASIR of IHBD cancers rose (0.43-1.84/100,000 population in males; 0.27-1.51 in females) but fell for EHBD (0.78-0.51/100,000 population in males; 0.62-0.39 in females). After transferring all 'Klatskin' tumours from IHBD to EHBD, there remained a marked increase in ASIR of IHBD cancers and a decrease in ASIR for EHBD, as only 1% of CC were reportedly 'Klatskin'. The US SEER data showed that ASIR for IHBD gradually rose from 0.59/100,000 population in 1990 to 0.91 in 2001, then sharply fell before plateauing at 0.60 by 2007. ASIR for EHBD remained relatively stable at around 0.80/100,000 population until 2001, then began increasing, to 0.97 by 2007. Annually, between 1995 and 2008, the vast majority of 'Klatskin' tumours in England and Wales were coded as IHBD. This was also the case in the SEER data until 2001, when the situation was reversed and subsequently most 'Klatskin' tumours were coded as EHBD. US trends coincide with a switch from ICD-O2 to ICD-O-3 in 2001. In the UK, the switch to ICD-O-3 only occurred in 2008. On questioning, cancer registries in England and Wales stated they would not code a CC described as 'hilar' with the designated 'Klatskin' histology code. If the tumour site is unspecified, most registries classify CC as intrahepatic.

Conclusions: Changes in ICD-classification may be influencing observed changes in IHBD and EHBD incidence rates. Coding misclassification is likely to have been skewing CC registration to an intrahepatic site, thereby contributing to the previously reported rise in intrahepatic tumours.