PerspectiveIn the Journals

Diabetics’ elevated risk for pancreatic cancer persisted long after diagnosis

Show Citation

August 20, 2014

A review of 15 case-control studies confirmed individuals with diabetes have an excess risk for pancreatic cancer.

Results showed a 30% excess risk for pancreatic cancer persisted for more than 20 years after diabetes diagnosis. That finding supports a causal role of diabetes in pancreatic cancer, according to researchers.

Cristina Bosetti, PhD, of the department of epidemiology at Istituto di Ricerche Faracologiche in Italy, and colleagues sought to assess individual-level data from case-control studies included in the Pancreatic Cancer Case-Control Consortium.

The studies included 8,305 cases and 13,987 controls. The researchers used multiple logistic regression models to estimate pooled ORs, and they made adjustments for relevant covariates.

Overall, 1,155 cases (15%) and 1,087 (8%) of controls reported a diagnosis of diabetes two or more years prior to cancer diagnosis (or interview, for controls), translating to an odds ratio of 1.9 (95% CI, 1.72-2.09).

Researchers observed consistent risk estimates when selected covariates — including BMI and smoking — were taken into account.

The risk for pancreatic cancer decreased with duration of diabetes; however, researchers observed a significant increased risk for at least 20 years after diabetes diagnosis (OR=1.3; 95% CI, 1.17-2.03).

Diabetics who used oral antidiabetics for ≥15 years demonstrated a decreased pancreatic cancer risk (OR=.31; 95% CI, 0.14-0.69). Among insulin users, researchers observed increased pancreatic cancer risk among those who reported short-term use (OR=5.6; 95% CI, 3.75-8.35 for less than 5 years) but not for those who reported longer duration of use (OR=.95; 95% CI, 0.53-1.7 for ≥15 years).

Disclosure: See the study for a full list of the researchers’ relevant financial disclosures.

itj+ Infographic

itj+ Perspective

PERSPECTIVE
Malcolm J. Moore

Malcolm J. Moore

The Pancreatic Cancer Case-Control Consortium, or PANC4, is a collaboration of a group of scientists of diverse biomedical disciplines from many different institutions across the world who conduct large case–control studies from pooled analyses of data to improve our understanding of pancreatic cancer. The size of the cohort of pooled data is a powerful tool for examining the epidemiology of pancreatic cancer, and the group has previously published important work examining the associations between tobacco use, alcohol consumption, pancreatitis and allergies with the risk for developing pancreatic cancer. The group continues and expands the scope of their work in this study looking at the influence of both diabetes and the use of antidiabetic medications on the risk for pancreatic cancer. There also is a nice discussion in the article that succinctly summarizes their work and places it into context. The PANC4 investigators demonstrate that a diagnosis of diabetes 2 or more years prior to cancer diagnosis (to censor out patients who develop diabetes just prior to diagnosis due to pancreatic insufficiency from the cancer) increases the risk for pancreatic cancer (OR=1.9), and that the chronic usage of oral antidiabetic agents is associated with a significantly reduced risk for development of pancreatic cancer (OR=0.31; 95% CI, 0.14-0.69). The latter finding is of particular significance given the interest in some oral antidiabetic agents as potential anticancer agents, through their effects in reducing insulin and insulin-like growth factor (IGF) levels — which can function to promote cell growth — or through a direct anti-proliferative effect. The results do indicate that further exploration of the potential anticancer effects of oral antidiabetic agents would be worthwhile.

Malcolm J. Moore, MD
HemOnc Today Editorial Board member

Disclosure: Moore reports no relevant financial disclosures.