• Pancreatic cancer
• Epidemiology
• early onset
• risk factor
• GWAS
• SNPs

Pancreatic ductal adenocarcinoma (PDAC) is a complex disease with an extremely poor prognosis, that typically affects people in their late adult life, however, in recent years, there has been a rising incidence among the younger population. These younger patients are defined as early-onset (EOPC) or very early-onset pancreatic cancer (VEOPC) depending on if they are diagnosed before 60 or 45 years of age. Due to their rarity, these subgroups of patients have been poorly studied and the cause triggering such an early onset is still unknown. Even though in the last years, several PDAC risk loci have been identified by genome-wide association studies (GWAS), almost nothing is known on the genetic background of EOPC and VEOPC. Several risk factors have been suggested to contribute to the development of PDAC, however, it is not clear if they are the same for younger patients. This study aims to identify specific PDAC age-onset genetic and non-genetic risk factors, which characterize the EOPC and VEOPC. With these purposes, an analysis at genome-wide scale was performed, using two-phase approach. In the discovery phase, conducted on 603 VEOPC cases and 6897 controls, from the Pancreatic Cancer Cohort Consortium (PanScan) and the Pancreatic Cancer Case-Control Consortium (PanC4), the association between almost 7.000.000 variants and VEOPC risk was assessed. In the second phase, the replication of variants identified in the first phase was attempted using an independent set of 309 VEOPC cases and 3325 controls, from the PANcreatic Disease Research (PANDoRA) consortium. Furthermore, for a subset of subjects from PANDoRA, the association between smoking, alcohol consumption, diagnosis of type 2 diabetes and PDAC risk was assessed across different age groups. Finally, to stratify the population by risk in the different onset ages a polygenic risk score (PRS) was generated including the 28 known PDAC risk loci and the ABO blood groups, in PANDoRA and PanScan + PanC4 subjects.
The results of the analysis showed six novel SNPs (SLC14A2-rs1381553, SEMA6A-rs3984967, CAMK1D-rs11257929, CNTNAP2-rs12671911, 1p22.2-rs72718310 and 18q12.3-rs111703463) that were statistically associated with VEOPC risk (p<1x10-4), but no statistically significant associations were observed in the replication phase. Furthermore, the results showed that smoking and diabetes affected risk regardless of the age of PDAC onset, showing both a trend effect along with the age categories, with a strong effect in younger subjects. For smoking, the ORs comparing current smokers to never-smokers were 2.85 (95% CI 1.65-4.90, p=1.63x10-4) for VEOPC, and 1.65 (95% CI 1.09-2.49, p=1.71x10-2) for individuals aged over 60 years. For diabetes, the ORs were 14.95 (95% CI 3.42-65.41, p=3.29x10-4) for VEOPC, and 2.48 (95% CI 1.80-3.43, p=3.67x10-8) for subjects aged over 60 years. The analysis of the PRS showed that subjects in the highest quintile (subjects with the major number of the risk alleles) had an increased risk of developing PDAC when compared with subjects in the lowest quintile, with similar results between youngest and oldest subjects. In conclusion, we did not find any novel variant associated with VEOPC risk, however, our study showed new evidence of smoking and diabetes as risk factors for PDAC, particularly with a greater effect in the early onset of PDAC.