Abstract
BACKGROUND: Cancer is a polygenic disease that involves the dysregulation of multiple biological events, including aberrant cell proliferation and evasion of immune responses, which collectively contribute to cancer development and progression. Although the polygenic risk score (PRS) has enabled assessment of the effects of cancer risk loci, the potential impact of other genetic variants with functional relevance to cancer biology remains largely unexplored. METHODS: We conducted a large-scale analysis of UK Biobank data, including 61,249 cancer cases and 254,203 controls of European ancestry across 16 cancer types. We assessed the impact of germline variants relevant to cancer gene expression and 33 cancer immune responses by constructing genetic scores (eQS: eQTL-specific genetic score; IMS: immune response-specific genetic score) and examining the utility of these scores independently and in addition to cancer-specific PRSs, for each of the 16 cancer types using multivariable logistic regression models. RESULTS: Here, we show that seven associations are significant after the Bonferroni correction (eQS of MAP3K1 and IDH2 in breast cancer, eQS of PAX8 in cervical cancer, eQS of NRG1 in thyroid cancer, IMS26 in lung cancer, IMS16 in prostate cancer, and IMS8 in testicular cancer). The IMS for interferon responses and effector T cells show protective associations with lung, prostate, and testicular cancers. Combining these IMS with cancer PRS improves risk stratification for testicular and prostate cancers. CONCLUSIONS: Our study provides insights into functional genetic scores associated with cancers.