Age-Dependent Interplay of Modifiable Risk Factors and Genetic Risk in Pancreatic Cancer

Importance: Identifying individuals at higher risk of pancreatic cancer (PC) is critical for developing targeted prevention strategies. Objective: To investigate whether the potential association of modifiable risk factors with PC varies across genetic risk profiles and age. Design, Setting, and Participants: This prospective population-based cohort study used data from participants in the UK Biobank, enrolled from 2006 to 2010 and followed up for a median (IQR) of 11.7 (10.9-12.9) years. Data analysis was conducted from January to September 2025. Exposures: A polygenic risk score (PRS) and a modifiable risk score (MRS; including smoking, excessive alcohol intake, unhealthy diet, physical inactivity, overweight, diabetes, and pancreatitis) for PC were each categorized as low, intermediate, and high. Main Outcomes and Measures: The main outcome, incident PC, was ascertained through record linkage with national registries. Hazard ratios (HRs) were estimated to assess the individual and joint associations of the MRS and PRS with PC across age groups (<60, 60-69, and ≥70 years). Standardized 10-year cumulative incidence (SCI) per 100 000 persons was calculated according to MRS and PRS strata across age groups. Results: A total of 290 645 participants (149 836 [51.6%] male; mean [SD] age, 56.1 [8.1] years) were included. During follow-up, 1187 participants (0.41%) developed incident PC. The PRS was consistently associated with increased PC risk across all age groups (HR per 1-SD increase in PRS: age <60 years, 1.39; 95% CI, 1.15-1.68; age 60-69 years, 1.66; 95% CI, 1.51-1.83; age ≥70 years, 1.37; 95% CI, 1.26-1.48; P for interaction = .43), while MRS showed stronger associations with PC in younger than older participants (HR per 1-point increase in MRS: age <60 years, 1.69; 95% CI, 1.43-2.01; age 60-69 years, 1.25; 95% CI, 1.14-1.36; age ≥70 years, 1.20; 95% CI, 1.12-1.29; P for interaction < .001). SCIs varied across combinations of PRS and MRS strata, and were higher with increasing MRS in all PRS groups. Among participants younger than 60 years, individuals with high PRS but low MRS had a lower SCI (38.2 events per 100 000 persons) than those with low PRS but high MRS (92.2 events per 100 000 persons). The absolute difference in SCIs between high and low MRS groups was greatest among individuals with high PRS, being 6.1-fold greater than that observed among those at low PRS (323.1 vs 52.8 per 100 000 persons). The corresponding fold differences were less pronounced in individuals aged 60 to 69 years and 70 years or older (1.2- and 2.8-fold; 500.1 vs 401.2 per 100 000 persons and 851.2 vs 304.9 per 100 000 persons, respectively). Conclusions and Relevance: This cohort study found that the associations of modifiable risk factors with PC were strongest among younger adults, especially those at high genetic risk. These findings highlight the importance of initiating prevention early in the life course, especially among individuals at high genetic risk.