Diabetes and the Risk for Biliary Tract Cancer
Abstract & Commentary
By Jerome W. Yates, MD
Hematology/Immunology Unit, National Institute on Aging, National Institutes of Health
Dr. Yates reports no financial relationships relevant to this field of study.
Synopsis:There has been much written recently about the association of diabetes and cancer. In this analysis of a large European cohort, the development of biliary tract and hepatocellular cancer was increased in diabetics compared with controls. However, the data as presented, though interesting, are not conclusive and additional research is called for to establish this association with confidence.
Source:Schlesinger S, et al. Diabetes mellitus, insulin treatment, diabetes duration, and risk of biliary tract cancer and hepatocellular carcinoma in a European cohort. Ann Oncol 2013;24:2449-2455.
Epidemiological studies suggest that individuals with diabetes mellitus (DM) are at higher risk of cancer.1,2Capitalizing on the large EPIC (European Prospective Investigation into Cancer and Nutrition) cohort including 363,426 participants with self-reported diabetes data, Schlesinger and coinvestigators conducted a prospective analysis to examine the risk of hepatic and biliary cancer among those with diabetes.3Multivariable adjusted relative risks (RR) and 95% confidence intervals (CI) were estimated from Cox regression models. In a nested, case-control subset, analyses were carried out in HCV/HBV-negative individuals.
During 8.5 years of follow-up, 204 biliary tract cancer (BTC) cases (including 75 gallbladder cancer [GBC] cases), and 176 hepatocellular cancer (HCC) cases were identified. Independent of body mass index and waist-to-height ratio, diabetes status was associated with higher risk of BTC and HCC (RR, 1.77; 95% CI, 1.00-3.13; and RR, 2.17; 95% CI, 1.36-3.47). For BTC, the risk seemed to be higher in participants with shorter diabetes duration and those not treated with insulin. Regarding cancer subsites, diabetes was only associated with GBC (RR, 2.72; 95% CI, 1.17-6.31). The risk for HCC was particularly higher in participants treated with insulin. The results were not appreciably different in HCV/HBV-negative individuals.
The authors concluded that their findings support the hypothesis that pre-existing diabetes is a risk factor for BTC (particularly GBC) and HCC, but that additional research will be essential to establish whether diabetes treatment or duration is associated with these cancers.
Indeed, multiple studies have indicated there is an association between DM and a variety of gastrointestinal cancers.1,2This paper focuses on the association with DM and BTC, GBC, and HCC. It reports the results of a prospective, nested, cohort study based on available data from the EPIC cohort relying on self-reported health, disease, and diet information from a large number of subjects (n = 363,426) for a mean follow-up time of 8.5 years. This study included information about general and abdominal obesity, information not available in many of the meta-analyses previously reported, and the authors felt this factor strengthened their results. They concluded that DM is a risk factor for BTC (RR, 1.77; 95% CI, 1.00-3.13), GBC (RR, 2.72; 95% CI, 1.17-6.31), and HCC (RR, 2.17; 95% CI, 1.36-3.47). The diagnosis of DM was based on self-report, which was only collected once with the initial questionnaire, and the only treatment data related to DM concerned insulin usage. BTC seemed to be more common in those with DM of short duration and HCC was increased in those treated with insulin.
Case control studies are often criticized because the cases and controls are not always selected from the same pool of subjects and unidentified confounding variables (associated both with the exposure of interest and an outcome of interest). These variables, if known, are optimally considered in the analysis. If unknown, they may obscure the study results. For this study, the authors considered factors such as obesity, gall stones, and hepatitis and controlled for these using appropriate statistical methodology.
The patients with DM not treated with insulin were at an increased risk for BTC, whereas only the participants treated with insulin and younger in age were at increased risk for HCC. These somewhat unexpected outcomes are of great interest but also raise questions about result validity. The highest relative risk for BTC was found in obese diabetics, a finding that clearly warrants additional research.
A major weakness of this study is the instability based on the small number of cancers of interest. Further, the diagnosis of DM was based on patient-supplied information at the time of study entry. Although the authors support the reliability of self-reported data in patients with DM, the multicultural background of participating subjects, the variability in their awareness of the diagnosis, and their understanding of the treatment may diminish that reliability. Because DM was determined only at the baseline (prevalence) and there was no reclassification of post-entry incident cases, three types of time biases were introduced (immortal, time window, and time lag).4Duration and severity of the DM is not consistently available, further blurring the estimates of the relationship with these cancers depending on the magnitude of the misclassification. In addition, there is no information about medications other than insulin, including metformin. Metformin appears to have a role in preventing some cancers.5This could decrease the number of cancers in the study population and decrease these relative risk estimates.
Thus, the findings are of interest, but design flaws limit the absolute validity of observations. Certainly, as the authors suggest, the association is of theoretical importance, but additional research is required to establish it with confidence.
- Coughlin SS, et al. Diabetes mellitus as a predictor of cancer mortality in a large cohort of US adults. Am J Epidemiol 2004;159:1160-1167.
- Giovannucci E, et al. Diabetes and cancer: A consensus report. Diabet Care 2010;33:1674-1685.
- Schlesinger S, et al. Diabetes mellitus, insulin treatment, diabetes duration, and risk of biliary tract cancer and hepatocellular carcinoma in a European cohort. Ann Oncol 2013;24:2449-2455.
- Suissa S, Azoulay L. Metformin and the risk of cancer: Time-related biases in observational studies. Diabet Care 2012;35:2665-2673.
- Del Barco S, et al. Metformin: Multi-faceted protection against cancer. Oncotarget 2011;2:896-917.