A risk factor is a characteristic or exposure that makes a disease more likely without being sufficient to cause it. For pancreatic cancer, the factors are usually divided into those that cannot be changed and those that can. None of them is a single decisive cause, and most people who develop the disease have none of the rare inherited conditions.
Fixed risk factors
Four factors cannot be changed: advanced age, male sex, African American race and a family history of pancreatic cancer. The first three mirror the age, sex and racial patterns of incidence.
Modifiable risk factors
Smoking is the most important modifiable factor. Relative risk compares the risk in an exposed group with that in a reference group; smokers have a relative risk of about 1.7, that is, a risk about 1.7 times that of the general population. The relation is dose-dependent, so the risk rises with the amount and duration smoked. Smoking is estimated to be at least one causative agent in more than 20–25% of cases. Alcohol use and obesity are also associated with the disease.
Diabetes has a more complicated relationship with pancreatic cancer than it first appears, because the direction of cause runs both ways. Long-standing diabetes raises the risk modestly, with a relative risk of roughly 1.7 to 1.9. A new diagnosis of diabetes, by contrast, is associated with a much higher risk — an 8-fold increase in the first 3 years — but this usually reflects the tumour rather than causing it: pancreatic cancer itself disturbs glucose metabolism, so new-onset diabetes can be an early clue to the cancer rather than a risk factor for it. About half of patients with pancreatic cancer already have diabetes when the cancer is found, and about 40% developed the diabetes in the 3 years beforehand.
Concern was raised that the incretin-based anti-diabetic drugs, the GLP-1 receptor agonists and DPP-4 inhibitors, might increase the risk of pancreatic cancer. Meta-analyses of randomised trials have not confirmed an association. Observational data suggest that metformin may be associated with a lower risk, but this has not been established as a protective effect.
These factors identify who is more likely to develop pancreatic cancer; how the cancer arises in the duct cells themselves is a separate question about the cellular changes involved.