Previous studies have suggested a “cholesterol-lowering effect” of preclinical pancreatic cancer, suggesting lower total cholesterol as a potential diagnostic marker. Leveraging repeated measurements of total cholesterol, this study aims to examine the temporal association of total cholesterol and pancreatic cancer incidence. We conducted a nested case-control study based on a Korean National Health Insurance Service−Health Screening Cohort, including 215 pancreatic cancer cases and 645 controls matched on age and sex. Conditional logistic regression was applied to estimate the odds ratio (OR) and 95% confidence interval (CI) for the associations of pancreatic cancer incidence with total cholesterol levels across different time windows over 11 years before pancreatic cancer diagnosis (recent, mid, distant). We found that, compared to participants with total cholesterol < 200 mg/dL in the recent 3 years prior to diagnosis, those having total cholesterol ≥ 240 mg/dL showed a significantly lower pancreatic cancer incidence (OR = 0.50 (0.27−0.93)). No significant association was found in relation to total cholesterol measured in the mid and distant past. When changes in total cholesterol over the three time periods were analyzed, compared with those with total cholesterol levels consistently below 240 mg/dL over the entire period, the OR of pancreatic cancer was 0.45 (0.20−1.03) for participants with recent-onset hypercholesterolemia, 1.89 (0.95−3.75) for recent-resolved hypercholesterolemia, and 0.71 (0.30−1.66) for consistent hypercholesterolemia. In conclusion, while high total cholesterol in the recent past may indicate a lower pancreatic cancer incidence, a recent decrease in total cholesterol may suggest an elevated incidence of pancreatic cancer.
Keywords: blood; incidence; lipid; pancreatic ductal adenocarcinoma; timing.