A Pleiotropic scan approach to discover new susceptibility loci for pancreatic ductal adenocarcinoma risk

Pancreatic ductal adenocarcinoma (PDAC) is the most common form of pancreatic cancer (PC). It is a relatively rare and lethal disease that represents the seventh leading cause of cancer death, and it was estimated to become the second by 2030. Through genome-wide association studies (GWAS), 27 common low-risk loci have been identified to be associated with PDAC risk. However, GWAS reported only the associations that reached the significant Pvalue threshold of 5×10-8, possibly leading to false negatives. A possible approach to uncover new variants is to study single nucleotide polymorphisms (SNPs) already known to be associated with at least one trait reported in previous GWAS. These SNPs are markers of human biological processes, they have a higher probability of association with other traits. Testing only these variants will reduce the number of tests, relaxing GWA significant threshold and increasing the power of the study. We call this approach “pleiotropic scan”. A pleiotropic scan has been already conducted for PDAC more than 10 years ago, but our knowledge on the pleiotropic variants that regulate multiple traits has greatly advanced. In addition, also the availability of data on PDAC SNPs has greatly increased. Our study was divided in three phases: in the first phase we downloaded the SNPs already known to be associated to any human trait at a genome-wide significance level from GWAS catalogue (186,829 SNPs); The next phase (discovery phase) the association of the selected SNPs was tested on 17,282 individuals (9563 cases and 7719 controls) belonging to the PanScan and PanC4 studies. At last, in the replication phase the promising SNPs (Pvalue < 0.05) were analyzed in three different populations Despite the large size of our study none of the variants reached the Pvalue threshold that we set for this study considering the correction for multiple testing (Pvalue= 1.34×10-6), however rs288762 (7q36.3) and rs4948550 (10q21.1) showed a lower Pvalue in the meta-analysis compared with the discovery phase. These two loci are interesting: rs288762 (7q36.3 locus) maps near the Sonic Hedgehog (SHH) gene, which was already observed to be overexpressed in cancers patients, moreover three genetic variants located in SHH gene are already known to be associated with the risk of developing PDAC, through our study we highlighting the importance of this region in the PDAC aetiology; the rs4948550) is located on chromosome 10 in the BicC Family RNA Binding Protein 1 (BICC1) gene which is involved in the negative regulation of the Wnt/beta-catenin (canonical) signaling. The altered activity of the Wnt signaling has been found associated with initiation, progression and therapeutic resistance in PDAC cases. Moreover, rs4948550 is in Linkage Disequilibrium (LD) whit rs4948317 (r2=0.82) associated with colorectal cancer (CRC). PDAC and CRC share several risk loci, our findings add to the knowledge of a potential overlap of the aetiopathogenesis of the two cancers. In conclusion, we have identified a potential novel pleiotropic region that merits further research especially in the form of fine mapping and ad hoc functional studies