Dysregulation of alternative splicing is implicated in many human diseases and understanding the genetic variation underlying transcript splicing is essential to dissect the molecular mechanisms of cancers. We aimed to provide a comprehensive functional dissection of splicing quantitative trait loci (sQTLs) in cancer and focus on elucidating its distinct role in colorectal cancer (CRC) mechanisms. Firstly, we performed a comprehensive sQTL analysis to identify genetic variants that control mRNA splicing across 33 cancer types from TCGA and independently validated in our 154 CRC tissues. Then, large-scale, multicenter, multiethnic case-control studies (34,585 cases and 76,023 controls) were conducted to examine the association of these sQTLs with CRC risk. A series of biological experiments in vitro and in vivo were performed to investigate the potential mechanisms of the candidate sQTLs and target genes. The molecular characterization of sQTL revealed its distinct role in cancer susceptibility. Tumor-specific sQTL further showed better response to cancer development. Besides, functionally informed polygenetic risk score highlighted the potentiality of sQTLs in the CRC prediction. Complemented by large-scale population studies, we identified that the risk allele (T) of a multi-ancestry-associated sQTL rs61746794 significantly increased the risk of CRC in Chinese populations (OR=1.20, 95% CI=1.12-1.29, P=8.82×10-7) and European populations (OR=1.11, 95% CI=1.07-1.16, P=1.13×10-7). rs61746794-T facilitated PRMT7 exon 16 splicing mediated by the RBP PRPF8, thus increasing the level of canonical PRMT7 isoform (PRMT7-V2). Overexpression of PRMT7-V2 significantly enhanced the growth of CRC cells and xenograft tumors, compared with PRMT7-V1. Mechanistically, PRMT7-V2 functions as an epigenetic writer that catalyzes the arginine methylation of H4R3 and H3R2, subsequently regulating diverse biological processes, including YAP, AKT, and KRAS pathway. A selective PRMT7 inhibitor, SGC3027, exhibited antitumor effects on human CRC cells. Our study provide an informative sQTLs resource and insights into the regulatory mechanisms linking splicing variants to cancer risk, serving as biomarkers and therapeutic targets.