Challenges in elucidating cholangiocarcinoma etiology.

HepatoBiliary Surg Nutr 2020 | http://dx.doi.org/10.21037/hbsn.2020.02.03 Cholangiocarcinomas (CCAs) are rare tumors that originate from cholangiocytes in the bile ducts and are classified as intraor extrahepatic (ICC or ECC). ICCs account for 10– 12% of liver cancers, while ECCs account for approximately one-third of biliary tract cancers (1). Incidence rates of CCA are geographically variable, with the highest rates in Asia. However, even in the highest incidence country (South Korea) the rates are only 2.8 and 2.2 per 100,000 personyears for ICC and ECC, respectively (2). Incidence rates of both ICC and ECC have been increasing in most countries globally, which potentially indicates a changing etiology (2). There are few established risk factors for CCA, including chronic conditions—primary sclerosing cholangitis and Caroli’s disease—and, in Asian countries, liver flukes and hepatolithiasis. However, the rarity of CCA makes elucidating the etiology of these tumors challenging. Compounding the challenges in studying CCA etiology is the evolving ICD coding over time, which has resulted in misclassification of perihilar or Klatskin tumors, which are ECCs arising proximal to the cystic duct (3). In response to these issues, we have conducted large, pooled analyses of prospective cohort studies—the Liver Cancer Pooling Project (LCPP) and the Biliary Tract Cancers Pooling Project (BiTCaPP). These pooling projects have allowed us to examine harmonized risk factors for ICC and ECC, with a focus on Western populations. In the BiTCaPP, we identified cigarette smoking and obesity, but not alcohol consumption, as etiologic factors that increase ECC risk (4,5). In the LCPP, we identified excess alcohol consumption, cigarette smoking, diabetes and obesity as ICC risk factors (6,7). Further, we conducted a systematic review and meta-analysis and reported that obesity and diabetes were associated with a 50% increased ICC risk (7). We only included studies with pre-diagnostic assessment of obesity and diabetes (i.e., cohort and nested case-control), as studies that assessed exposure at or after time of cancer diagnosis are susceptible to reverse causation, as ICC patients often present with cachexia. In the January issue of the Journal of Hepatology, Clements et al. conducted a systematic review and meta-analysis of risk factors for ICC and ECC (8). To conduct this study, the authors searched a singular source (MEDLINE) to identify published articles. They reported that bile duct cysts and stones, chronic hepatitis B and C viral infections, and inflammatory bowel disease increased risk of ICC and ECC 2to 35-fold. More modest increased risk of ICC and ECC were noted for alcohol, smoking, and diabetes, while no associations were reported for hypertension or obesity (8). The objective of systematic reviews and meta-analyses are to rigorously review and synthesize the entire body of scientific studies—that is, to determine if there is replication in the published literature. The study by Clements et al. is the largest and most comprehensive meta-analysis to date— examining 13 potential risk factors in 25 studies of ICC and ECC (8). However, producing an averaged effect estimate across all studies may come at the expense of identifying and reporting heterogeneity in study-specific estimates, which can illuminate why certain studies did not replicate prior findings. The authors present the individual forest and funnel plots, which show minimal publication bias and homogeneous results for most exposures. However, the study-specific estimates are not consistent for smoking, alcohol, hypertension, and obesity. The cautious reader should apply the same scrutiny to Commentary