Transcriptome-wide association analysis of survival in colorectal cancer patients

Colorectal cancer (CRC) is the second most common cause of cancer death in the UK with 15,900 CRC deaths occurring in 2014. The risk of recurrence and death depends very much on the stage of disease at the time of diagnosis with 5-year survival rate >90% for stage I and <10% for stage IV. However, even individuals with the same surgical stage of disease may have substantial different prognosis and risk of recurrence. Hence, prediction at the individual patient level is very poor for survival outcomes, likelihood of locoregional recurrence or metastatic spread, even within any given tumour stage. There is a pressing need for new approaches to guide clinical management and target treatment to those most likely to gain benefit. Precision medicine absolutely requires such predictive performance to be effective.

Recent analysis of all genes expressed (transcriptome) in tumours from 597 individuals with CRC using an RNA sequencing (RNA-seq) approach has identified 593 genes associated with survival. However, little is known about the association between the transcriptome of normal colonic tissue and cancer prognosis and survival. Published evidence does suggest that the transcriptome profile of normal tissue can contribute to the prognosis prediction. Recent progress in our understanding of gene expression and its relationship to genetics made it possible to predict and impute gene expression level into genome-wide data and perform transcriptome-wide association studies (TWAS) to measure the strength of association between gene expression and outcomes of interest.

Here we proposed to perform a TWAS of CRC survival in UK Biobank and Scottish Colorectal Cancer Study (SOCCS). The overarching aim is to generate data to develop and test new predictive algorithms for survival outcome, disease relapse patterns and potentially even with the ability to identify targets for individualised chemotherapy.