Pancreatitis and pancreatic cancer in two large pooled case–control studies

OBJECTIVE: The association between duration of pancreatitis and pancreatic cancer has not been well characterized in large population-based studies. We conducted detailed analyses to determine the association between pancreatitis onset and pancreatic cancer risk. METHODS: Data from two case–control studies of pancreatic cancer (n = 4515) in the San Francisco Bay Area and the M.D. Anderson Cancer Center were pooled for analysis (1,663 cases, 2,852 frequency-matched controls). Adjusted odds ratios (OR) were estimated using a random-effects model. RESULTS: In the pooled multivariable model, history of pancreatitis was associated with a 7.2-fold increased risk estimate for pancreatic cancer [95% confidence interval (CI): 4.0, 13]. The risk estimate was nearly 10-fold in participants aged <55 years (OR = 9.9, 95% CI: 3.5, 28). A shorter temporal history of pancreatitis was more closely associated with pancreatic cancer than was a longer temporal history: <3 years (OR = 29, 95% CI: 12, 71), 3–10 years (OR = 2.6, 95% CI: 1.5, 5.6), and >10 years (OR = 1.8, 95% CI: 0.7, 4.5, p (trend) < 0.001). CONCLUSION: A short temporal history of pancreatitis was highly associated with pancreatic cancer, suggesting that pancreatitis may be an early manifestation of pancreatic cancer in some individuals. Pancreatic cancer should be considered in the differential diagnosis of individuals with an episode of pancreatitis.