Identifying risk factors for pancreatic cancer in a large Japanese cohort using an insurance and health check-up database

database[Title] 2026-04-17

Pancreatology. 2026 Apr 6:S1424-3903(26)00154-7. doi: 10.1016/j.pan.2026.04.012. Online ahead of print.

ABSTRACT

BACKGROUND: Pancreatic cancer (PC) has a poor prognosis, with most cases diagnosed at an unresectable stage. Identifying high-risk individuals within the general population is essential for targeted surveillance and early detection. We aimed to clarify the contributions of established and potential risk factors to subsequent PC occurrence.

METHODS: We conducted a retrospective population-based cohort study using the Shizuoka Kokuho Database, which links administrative claims with annual health checkup data. Adults with at least one health checkup and a 1-year baseline period were included. Incident PC was identified using ICD-10 codes. Demographic, health check, lifestyle, and comorbidity-related variables were evaluated using cause-specific Cox proportional hazards models, with subgroup analyses by sex and age.

RESULTS: Among 641,979 individuals in the analytical cohort, 4313 developed PC during a median follow-up of 6.8 years. PC risk increased steeply with age and was higher in men than in women. HbA1c, rather than diagnosed diabetes, showed a clear dose-response association with PC risk, reaching an HR of 2.85 (95% CI 2.29-3.56) at HbA1c ≥ 8%. Chronic pancreatitis (HR 1.71, 95% CI 1.32-2.22), pancreatic cystic disease (HR 4.72, 95% CI 3.59-6.20), hypertension, deficiency anemia, aspartate aminotransferase elevation, and habitual smoking were independently associated with an increased risk of PC, whereas dyslipidemia and high low-density lipoprotein cholesterol levels were inversely associated. Metabolic and lifestyle-related risk factors were more evident in individuals aged ≥60 years.

CONCLUSION: Our findings may help prioritize high-risk groups for targeted surveillance and inform the development of risk-stratified early detection strategies for PC.

PMID:41963187 | DOI:10.1016/j.pan.2026.04.012