Familial and second primary pancreatic cancers: a nationwide epidemiologic study from Sweden.

Abstract

Familial risk of pancreatic cancer has been mainly assessed through case-control studies based on reported but not medically verified cancers in family members. We used the nationwide Swedish Family-Cancer Database on 10.2 million individuals and 21,000 pancreatic cancers to calculate standardized incidence ratios (SIRs) and 95% confidence intervals (CIs) for pancreatic cancer in 0- to 66-year-old offspring of parents with pancreatic or other specified tumors. Additionally, SIRs for second primary pancreatic cancers were analyzed after any first neoplasm. SIRs for pancreatic cancer (1.68, 95% CI 1.16-2.35) and pancreatic adenocarcinoma (1.73, 95% CI 1.13-2.54) were increased when a parent presented with pancreatic cancer. The risk was not dependent on diagnostic age of offspring or parents. Pancreatic cancer was associated with parental lung, rectal or endometrial cancer and with melanoma. SIRs for pancreatic cancer were 10.01 and 7.96 among offspring who were diagnosed before age 50 years when parents were diagnosed with squamous cell and adenocarcinoma of the lung, respectively, before age 60 years. The population-attributable proportion of familial pancreatic cancer was 1.1%. Risks for second pancreatic cancers were increased in men and women after small intestinal, colon and bladder cancer. The degree of familial clustering for pancreatic cancer and its population-attributable proportion were lower than the data cited in the literature. Clustering of pancreatic cancer with sites presenting in hereditary nonpolyposis colorectal cancer was noted. The strong association of pancreatic and lung cancers is puzzling, and it remains unclear to what extent this represents familial sharing of smoking habits.