Identification of potential drug targets for four site-specific cancers by integrating human plasma proteome with genome

Drug targets supported by genetic evidence with a several-fold higher probability of success in clinical trials. We performed a comprehensive proteome-wide Mendelian randomization (MR) analysis to identify causal proteins and potential therapeutic targets for four site-specific cancers. A total of 13,248 protein quantitative trait loci for 4853 plasma proteins were utilized for proteome-wide MR analysis. Identification of cancer causal proteins in the discovery cohort and further validation in the replication cohort. Colocalization, summary-data-based MR (SMR) analysis, and transcriptome-wide association studies (TWAS) were performed to check the accuracy of the candidate proteins. Two-step MR analysis was used to explore the effects of plasma protein-mediated 248 modifiable factors on cancer. Phenome-wide MR (Phe-MR) analysis, druggability evaluation, and single-cell type expression analysis further assessed the potential of causal proteins. Combining the results of the meta-analysis of MR estimates from the two cohorts, 21, 2, 24 and 1 causal proteins were identified in breast, lung, prostate and stomach cancers, respectively. Evidence from colocalization, SMR analysis, and TWAS highlighted CD36, DNPH1, and PLXND1 as the most promising drug targets for breast cancer, and ZNF175 for prostate cancer. 1 new potential biomarker (PLXND1) for breast cancer, 2 new promising targets (RELL1, DEFB119) for lung cancer, and 8 new circulating biomarkers (ARFIP2, CCN6, CTRB2, HTR7, MRPL33, TNFRSF6B, VAMP5, ZNF175) for prostate cancer were firstly reported. Some plasma proteins may mediate the association of these cancers with other systemic diseases. Additionally, genetically predicted higher BMI and overweight may reduce breast cancer risk by altering CASP8, ADM, PLXND1, TNFRSF9, ULK3 and VSIG4 protein levels. Causal proteins of breast and prostate cancer were expressed predominantly on macrophages in cancerous tissues. This study genetically identified several cancer causal proteins which provided new perspectives for the understanding of the etiology and development of novel targeted drugs for cancer.