Approved Research

Predicting gastrointestinal tract cancers using genes, biomarkers and clinical risk factors

Lay summary

Specifically, we aim to:

1. Identify risk factors for digestive tract cancers that could be useful for predicting a person's risk of developing a digestive tract cancer.
2. Developing algorithms/models using a combination of their lifestyle risk factors, biomarkers and genes to predict a person's risk of developing digestive tract cancers within a few years (which might suggest a pre-cancerous lesion is present already).
3. Developing algorithms/models using a combination of their lifestyle risk factors, biomarkers and genes to predict a person's risk of developing digestive tract cancers within a few years (which might suggest a pre-cancerous lesion is present already), in individuals with medical conditions associated with risk of cancer.

Digestive tract cancers (such as pancreatic cancer) often have poor survival, due to late diagnosis. Identifying individuals at a high risk of these cancers may allow screening to be provided. Use of a combination of risk factors like smoking, biomarkers and genes may help to identify higher risk individuals, who could benefit from screening.

Linkage between UK Biobank and cancer registries will provide information on cohort members diagnosed with digestive tract cancers. Using appropriate statistical techniques, we will use that information to develop algorithms to predict how likely an individual was to develop a specific digestive tract cancer based on their lifestyle, biomarkers and genetic information. We will then assess how well these models predict an individual's risk of cancer, and how well they distinguish between people who did develop cancer and people who stayed cancer-free.

In particular, the large size of the UK Biobank will facilitate the exploration of risk factors according to cancer subtypes, which could lead to better tailoring of risk prediction which could be used to inform screening decisions.