Using human proteomic and lipidomic to exploration drug target of bladder cancer

The goal of this study is to explore the genetic predictive associations between plasma proteomics and bladder cancer using the Mendelian randomization (MR) method, and to investigate whether plasma lipidomics characteristics mediate these associations.

The genome-wide association study (GWAS) data for plasma proteins from deCODE Genetics included 4,907 aptamers for plasma proteins in 35,559 Icelandic individuals. Lipidomic GWAS data of 179 traits (GCST90277238-GCST90277416) were extracted from the GWAS Catalog, and bladder cancer GWAS summary statistics were sourced from FinnGen R8 (2,380 cases and 259,583 controls). Two-sample MR evaluated the causal links between genetically predicted plasma proteins and bladder cancer risk; mediation MR further examined mediating roles of lipidomic. Primary causal analyses adopted the MR Wald ratio and inverse variance-weighted approaches. Heterogeneity was tested by the Cochrane Q test, horizontal pleiotropy by MR-Egger intercept test, and leave-one-out analysis was used for sensitivity assessment.

Our findings suggest that genetically predicted levels of 29 proteins may be associated with the risk of bladder cancer. Among these, 17 proteins showed inverse associations with bladder cancer risk, while 12 proteins showed positive associations. A negative mediation proportion indicates that the estimated direction of the indirect effect is opposite to that of the total effect. For example, the mediation proportion of -56% for ETFA suggests that the estimated indirect effect through GCST90277318 is in the opposite direction to the overall association observed between genetically predicted ETFA levels and bladder cancer risk. When ADGRF1 was considered as the exposure and GCST90277318 as the mediator, the estimated proportion mediated was 25%. For NDUFB4 and GCST90277318, the estimated proportion mediated was 20%. For ETFA and GCST90277318, the estimated proportion mediated was -56%, whereas the estimated mediation proportions for most other protein-lipid pairs were approximately 10%.These results reflect putative causal estimates derived from genetic instruments rather than direct causal effects.The findings should be interpreted in light of several limitations, including the use of summary-level data, predominantly European ancestry populations, and the potential bias from lifestyle and clinical confounders,and the need for experimental validation of the identified associations and proposed biological mechanisms.

This MR study revealed suggestive genetically predicted associations between 29 plasma proteins and bladder cancer risk, and implied that plasma lipidomic traits may partially mediate a subset of these relationships. These observations may shed light on potential molecular mechanisms relevant to bladder cancer and offer candidate proteins for further functional research and therapeutic target verification.