Dong Hang

Ph.D.

Nanjing Medical University

Associate Professor

PLCO (Learn more about this study)

PLCO-793

Jun 21, 2021

Integrating genetic susceptibility, blood biomarkers, and traditional factors to develop predictive models for colorectal adenoma and cancer

Identification of colorectal cancer (CRC) at an early stage and interrupting the natural history can significantly reduce CRC incidence and mortality. Although colonoscopy is regarded as the gold standard of CRC screening, it is an invasive procedure requiring a high level of expertise. Stratifying the population into different risk categories can potentially improve the efficiency of screening and facilitate the development of tailored intervention strategies for CRC prevention.
Most of the previously developed models for predicting CRC risk are based on risk factors derived from questionnaires and have shown a modest discriminatory ability. Genetic susceptibility contributes to CRC risk and approximately 100 variants have been identified in genome-wide association studies. Models including genetic variants alone or in combination with risk factors show a C-statistic of 0.6-0.8 in some studies but lack validation. In addition, several groups of blood biomarkers have been implicated in the early stage of colorectal carcinogenesis, including those related to inflammation, metabolic disturbances, and sex hormones. However, the prediction performance of models including blood biomarkers remains largely unknown. Therefore, there is an urgent need to examine the effect of combining genetic susceptibility, blood biomarkers, and traditional risk factors for CRC prediction. We will develop and validate models for CRC and adenomas, which incorporate genetic risk scores, blood biomarkers, and traditional risk factor scores, based on the UK Biobank and PLCO resources, as well as our in-house cohort data.