Using human genetics to identify novel biomarkers for enhanced prediction and early detection of cancer

Cancer is the 2nd leading cause of death globally. Despite important advances in treatment in recent decades, the prognosis for many cancers remains poor because individuals typically present with advanced disease at which point curative treatment is no longer an option. The NHS Long Term Plan aims to increase the percentage of cancers caught at an early stage, thus potentially curable, from 50-75% by 2028. Improvements in early detection and prediction of cancer is an important vehicle to achieve that goal. Proteomic Reverse Mendelian randomization (RMR) is a novel human genetics approach that uses germline genetic variants to identify circulating proteins associated with a genetic tendency to develop disease. The approach can be used to identify circulating proteins implicated in disease aetiology as well as non-causal "bystander" proteins that may nonetheless be informative biomarkers for disease prediction and detection. In this project, the student will use a Proteomic RMR framework to identify proteins associated with a genetic predisposition to develop cancer. These findings will be validated using direct protein measures and cancer data in UK Biobank (UKB). The potential clinical application of the identified proteins will be assessed by integrating circulating protein data into current clinical cancer risk prediction models, using linked medical records in UKB.